Introduction to Deep Technologies
PREVENT Project
Start
Definition and scope of deep technologies
"Deep technology" encompasses advanced technologies that tackle complex global issues such as climate change and healthcare with innovative solutions. Unlike traditional technologies, deep tech focuses on addressing local problems proactively and efficiently. Key areas of deep tech include Artificial Intelligence, Space Tech, and Computational Biology, among others. These technologies are distinguished by their disruptive potential and reliance on cutting-edge scientific breakthroughs.
Index
Definition of Deep Technologies
Remote Sensing
Robotics
Artificial Intelligence
Computer Vision
IoT and communications
Objectives
Through this module, learners will gain the role of advanced technologies in disaster prevention and mitigation. By exploring real-world case studies, scientific research, and interactive simulations, students will develop the ability to classify different types of natural disasters, analyze their social, economic, and environmental consequences, and evaluate the impact of climate change on disaster frequency and severity. Furthermore, learners will acquire critical thinking skills to assess disaster risk management strategies, understand how deep technologies such as AI, IoT, and satellite imaging contribute to early warning systems, and explore innovative disaster resilience frameworks.
By the end of the module, students will be able to apply knowledge in risk assessment, disaster response planning, and climate adaptation strategies, equipping them with essential skills for careers in environmental science, emergency management, and sustainable development.
'True learning begins when knowledge inspires action, and understanding drives change. Equip yourself to turn challenges into opportunities.'
'Harnessing the power of new technologies, we can predict, prepare for, and mitigate the impact of natural disasters—turning technology into a lifeline for our future.'
01
Definition of Deep Technologies
01
Definition of Deep Technologies
Deep Technologies refer to cutting-edge innovations that are based on substantial scientific and engineering advancements. These technologies often require significant research and development (R&D) and have the potential to create disruptive, transformative changes across various industries. Unlike incremental technological improvements, deep tech innovations are rooted in fundamental scientific breakthroughs and complex engineering. Key Characteristics of Natural Disasters:
- Science-Driven Innovation – Deep tech is built on foundational scientific and engineering principles, often originating from fields like artificial intelligence (AI), robotics, nanotechnology, biotechnology, quantum computing, and advanced materials.
- High R&D Intensity – These technologies demand extensive research, testing, and investment before reaching commercial viability.
- Cross-Disciplinary Nature – Many deep tech innovations integrate multiple scientific fields, such as AI combined with climate modeling or biotechnology with advanced computing.
Deep Tech in the Context of Natural Disasters
When combined with natural disaster preparedness, response, and mitigation, deep technologies can offer groundbreaking solutions. Examples include:
- AI & Machine Learning – Predicting disasters (earthquakes, floods, hurricanes) with advanced modeling.
- Remote Sensing & Drones – Rapid assessment of disaster zones and monitoring environmental changes.
- Blockchain – Secure and transparent aid distribution during crises.
- Quantum Computing – Enhanced climate simulations for better disaster prediction.
- Advanced Materials – Earthquake-resistant construction materials.
- IoT & Smart Sensors – Real-time monitoring of environmental risks and early warning systems.
02
Artificial Intelligence
02
Introduction to AI
Artificial Intelligence (AI) is a branch of computer science focused on creating systems that can perform tasks traditionally requiring human intelligence. These tasks include learning, reasoning, problem-solving, perception, language understanding, and decision-making. AI systems can analyze vast amounts of data, recognize patterns, and improve their performance over time through machine learning algorithms.
AI is classified into different categories based on capabilities and functions.
The primary types include:
- Narrow AI (Weak AI): AI systems designed to perform a specific task, such as facial recognition, speech translation, or medical diagnosis. These are the most commonly used AI applications today.
- General AI (Strong AI): A theoretical AI system capable of understanding, learning, and performing any intellectual task that a human can do. This level of AI is still under research and development.
- Super AI: A hypothetical future stage where AI surpasses human intelligence in all aspects, including creativity, problem-solving, and decision-making.
AI technologies rely on various subfields, including:
- Machine Learning (ML): Algorithms that enable computers to learn from data without being explicitly programmed.
- Deep Learning: A subset of ML that mimics human neural networks to process complex patterns in data.
- Natural Language Processing (NLP): The ability of AI to understand and generate human language, as seen in virtual assistants and chatbots.
- Computer Vision: The ability to analyze and interpret visual data from images and videos.
AI applications
Generic Applications
Business Intelligence (BI)
Observation systems
Healthcare
Education
Manufacturing
Finance
Other Applications
03
Remote Sensing
03
Introduction to Remote Sensing
Remote sensing describes the collection of data about an object, area, or phenomenon from a distance with a device that is not in contact with the object”. Remote sensing is generally broadly defined as the acquisition of data and information about Earth's surfaces and objects without physical contact. Its exact definition is challenging and definitions like the one above is often considered too general. There are many ways to define remote sensing in the literature. Two common elements characterize this scientific field a) the concept of "information collection" and b) "observation from a distance"
Types and causes of Floods
Remote sensing has gone through many stages to reach its current level. It is a quite evolutionary scientific field that relies on various other scientific disciplines such as mathematics, physics, computer science, etc. Since the discovery of photography in the 19th century, significant progress has been made in environmental remote sensing. Although it is quite difficult to precisely define the starting point of remote sensing and its evolutionary phases, five main stages are distinguished during its development . While Remote Sensing (RS) is its own field, it often acts as a complement to GIS analyses, adding unique information and analysis techniques. There are two types of RS, active and passive and they are generally used for different applications. Active RS involves sending out a signal and waiting for its return to the sensor. RADAR and LIDAR are examples of active RS, as they send out energy, microwave and laser pulses respectively and record the signals as they bounce back.
Remote Sensing Disadvantages
Remote Sensing Advantages
03
Remote Sensing for Natural Disaster Monitoring and Mitigation
Soil moisture is a critical variable for understanding water resources and natural hazards such as floods and landslides. Accurate estimation of spatial and temporal variations in moisture is important for environmental studies and improving flood forecasts, especially in medium and large drainage basins where floods often cause disasters. The condition of surface soil moisture is critical for assessing rainwater infiltration or runoff, so this information is vital for flood prediction models. Moreover, soil moisture in mountainous areas is a key factor for landslides. Since the 1970s, research in this field has used techniques that cover the entire electromagnetic spectrum, from optical to microwave. These techniques vary in terms of wavelength range, source of electromagnetic energy, sensor response and the physical relationship between response and soil moisture. The use of solar radiation measures the reflected sunlight from the Earth's surface. However, microwaves and thermal infrared have been used more frequently for calculating soil moisture. The use of the thermal section is based on measuring the surface temperature of the soil and methods such as thermal inertia and the temperature/vegetation index. Thermal inertia is simple and accurate in areas with minimal or no vegetation. In summary, remote sensing through active and passive sensors provides valuable data for estimating soil moisture, improving natural hazard prediction and water resource management.
04
Robotics
04
Introduction to Robotics
Robotics, a multidisciplinary field at the intersection of engineering, computer science, and artificial intelligence, plays an increasingly vital role in addressing the complex challenges of natural disaster management. Robots are designed to perform tasks that are dangerous, repetitive, or otherwise unsuitable for humans, making them invaluable in disaster scenarios. The evolution of robotics, driven by advancements in sensors, machine learning, and autonomous systems, has expanded their capabilities and applications, enabling more effective and efficient responses to natural disasters.
In disaster management, robotics can perform a variety of crucial functions, from search and rescue operations to environmental monitoring and infrastructure inspection. Autonomous drones, for instance, can quickly survey affected areas, providing real-time data and high-resolution imagery to aid in decision-making and coordination efforts. Ground robots can navigate hazardous terrains to locate survivors, deliver supplies, and perform structural assessments. Furthermore, underwater robots are essential for operations in flooded or submerged environments.
Different categories of robotics in disaster management Search and Rescue Robots – Assist in locating and saving survivors in collapsed structures or hazardous zones. Firefighting Robots – Suppress wildfires and prevent fire spread using AI and automated water or foam dispersal. Aerial Drones – Provide real-time surveillance, assess damage, and identify hazards from above. Underwater and Amphibious Robots – Aid in flood rescues and assess underwater damage. Inspection and Recovery Robots – Evaluate structural integrity and assist in clearing debris post-disaster.
Robotics aids disaster response by enhancing rescue, fire suppression, and damage assessment with AI and sensors.
04
History of Robotics
Robotics has significantly advanced disaster response, from early industrial automation to AI-powered autonomous systems. Robots assist in search and rescue, hazard assessment, and crisis recovery. Key deployments include Chernobyl (1986), 9/11 (2001), and Fukushima (2011), where robots navigated hazardous environments to support emergency teams. Today, AI-driven drones, autonomous vehicles, and robotic systems enhance real-time data collection, structural analysis, and aid delivery. As technology evolves, robotics continues to improve disaster management, making responses faster, safer, and more effective.
05
Iot and Communications
05
Introduction to IoT and communications
Internet of Things has become a very popular topic of research and Innovation mainly due to the ubiquitous transformation of computing. Physical devices have become “smart”, being able to sense, communicate in a pervasive way and interact with their environment offering useful applications and solutions to humankind. Today they find applications in a wide range of activities like Health, Transportation, Agriculture, Home and Industrial Automation, Retail and many more. It is expected an exponential growth of network connections which should be facilitated by powerful networks.
Fundamental Characteristics of the IoT and Communications
Interconnectivity: Any IoT device can be interconnected with the global Information and Communication Infrastructure.
Things-related services: The IoT provisions services which concern the connected “things” within their constraints such as privacy protections and semantic consistency between physical things and their associated virtual things.
Heterogeneity: Heterogeneous IoT devices with different hardware and networking characteristics get connected and interact with other devices or platforms on various types networks.
Dynamic Changes: While roaming and interacting in an IoT system, devices change their state dynamically. For example, sleeping and waking up, get connected or disconnected while changing their location and speed. Additionally and equally important the number of connected devices changes dynamically.
Enormous scale: Usually the number of devices that need to be managed and that of the devices that communicate with each other is significantly larger than the ones that are connected to the Internet. This practically means that the underlying communication network fabric needs to be able to support the high volume of data and the rate and quality that this needs to be exchanged.
Aspects of IoT
Useful Definitions
IoT Architecture
Observation systems
Fundamental Characteristics
IoT Requirements
IoT Devices and Components
IoT Challenges
IoT Challenges
Natural Disaster IoT Applications
Real-Time Data Collection Collecting real-time data is crucial for disaster prediction, detection, and management. Sensors measuring air quality, humidity, and temperature help authorities assess risks and take preventive measures. Interconnected Devices IoT systems integrate diverse devices with different capabilities for resource allocation and strategic response in disasters. Early Warnings & Predictions AI and machine learning analyze real-time data to predict disasters, enabling early intervention and saving lives. Data Sharing & Analysis Effective disaster management requires collaboration among first responders, authorities, and the public through data sharing. Human-less Interaction Drones and robots aid disaster response by performing remote monitoring, search, and intervention in hazardous areas. Smart Infrastructure & Resilience IoT-enabled buildings with sensor networks enhance resilience by monitoring structures, shutting down utilities, and activating fire suppression systems to reduce disaster impact.
06
Computer Vision
06
Computer Vision
Computer vision enables machines to interpret and analyze visual data, automating tasks like object detection, image segmentation, and 3D scene reconstruction. Advances in AI, computational power, and deep learning, particularly CNNs, have revolutionized the field, making it integral to applications such as autonomous vehicles, facial recognition, and augmented reality. In disaster management, computer vision enhances response teams by enabling real-time monitoring, search and rescue, and damage assessment. By providing accurate visual data, it aids decision-making during emergencies. This section aims to help educators understand computer vision, its role in disaster response, and how to integrate it into education, preparing students to use these technologies for crisis preparedness.
06
History of Computer Vision
Computer vision emerged in the 1960s-70s, focusing on 2D image analysis and basic shape recognition. The 1980s introduced motion detection, optical flow, and industrial applications. In the 1990s, machine learning techniques like SVM and PCA improved face recognition and object detection. The 2000s saw the rise of CNNs, revolutionizing image analysis with deep learning, highlighted by AlexNet (2012). The 2010s brought advanced architectures (VGG, ResNet) and real-world applications like autonomous vehicles and medical imaging. In the 2020s, innovations like GANs, transformers, and self-supervised learning push the field forward. Generative AI enables image synthesis, data augmentation, and AI-driven simulations, enhancing training and real-world applications. Computer vision now plays a crucial role in disaster management, healthcare, and environmental monitoring, solving global challenges.
Thank you!
Useful Definitions
Device: In the IoT context, this is a piece of equipment must be able to communicate and could optionally sense, act, capture data, store data or process data. Its only mandatory capability is the communication.
Thing: An object inside the IoT system which is capable of being identified and integrated into communication system.
Physical Thing: An object of the physical world which is able of being sensed, actuated and connected is known as a physical thing (e.g. industrial robots, electrical equipment etc.)
Virtual Thing: An object in the information world capable of being stored, processed and accessed is known as virtual thing. For example, multimedia content, application software etc.
Internet of Things: A global information infrastructure which enables advanced services by interconnecting Things (Physical and/or virtual) based on existing and/or evolving interoperable technologies. The IoT includes functions for identification, data capture, processing, and communication to offer different kinds of applications whilst ensuring security and privacy.
Fundamental Characteristics
Interconnectivity: Any IoT device can be interconnected with the global Information and Communication Infrastructure.
Things-related services: The IoT provisions services which concern the connected “things” within their constraints such as privacy protections and semantic consistency between physical things and their associated virtual things.
Heterogeneity: Heterogeneous IoT devices with different hardware and networking characteristics get connected and interact with other devices or platforms on various types networks.
Dynamic Changes: While roaming and interacting in an IoT system, devices change their state dynamically. For example, sleeping and waking up, get connected or disconnected while changing their location and speed. Additionally and equally important the number of connected devices changes dynamically.
Enormous scale: Usually the number of devices that need to be managed and that of the devices that communicate with each other is significantly larger than the ones that are connected to the Internet. This practically means that the underlying communication network fabric needs to be able to support the high volume of data and the rate and quality that this needs to be exchanged.
Business Intelligence (BI)
AI-powered BI tools enhance data collection, analysis, and visualization, improving decision-making and productivity while reducing costs. Data Collection: Gathering structured (e.g., databases) and unstructured data (e.g., text, images).
Data Analysis: Identifying patterns, trends, and relationships in data.
Data Visualization: Creating visual representations for easier comprehension.
Decision-Making: Providing insights and recommendations to support data-driven decisions.
Education
AI can personalize learning, enhance student engagement, and automate administrative tasks:
Personalized Learning: Adapting learning experiences based on student progress.
Improved Engagement: Providing interactive learning experiences and real-time feedback.
Administrative Tasks: Automating tasks like grading and scheduling to free up teacher time.
Generic Applications
Natural Language Processing (NLP): NLP enables computers to understand and generate human language. Applications include machine translation, spam filtering, and sentiment analysis.
Computer Vision: Computer vision allows computers to interpret visual content, used in applications like self-driving cars, facial recognition, and object detection.
Machine Learning (ML): ML enables computers to learn from data and improve performance over time. Applications include predictive analytics, fraud detection, and recommendation systems.
Robotics: Robotics involves designing, constructing, and operating robots for applications such as manufacturing, healthcare, and space exploration.
Manufacturing
AI improves manufacturing efficiency, productivity, and quality control:
Improved Efficiency: Automating tasks such as assembly and inspection.
Increased Productivity: Optimizing production processes.
Quality Improvement: Detecting defects and enhancing quality control.
IoT Devices and Components
Regarding the processing capabilities devices are classified as:
Devices with no processing capability: Passive devices, usually low-cost and with no microcontrollers. A typical example is an RFID.
Devices with low processing capabilities: Their processing capabilities are limited to the reading and writing data from or to sensors and actuators and sending this data to IoT applications, but they are not able to make decisions or run complex algorithm. They are typically low cost and usually embed a very low-power and low-cost microcontroller. A typical example is a smart light or a door sensor.
Devices with high procession capabilities: They have enough processing power to enable them to make decisions and run complex algorithms. They are typically high cost as they employ a powerful microcontroller. (e.g. a smart cooling system, or a smart thermostat)
Regarding the connectivity capabilities devices can be classified as:
Devices with low connectivity: This kind of devices do not connect directly to the communication network to transfer the data but instead they rely on additional elements (e.g. gateway) to perform communications tasks (e.g. protocol translation or internet connectivity).
Devices with High connectivity: They have the hardware and ability to directly connect to the network to transfer the data.
IoT Architecture
A physical thing can be mapped (or represented) by one or more virtual things in the Information domain. Information is being collected by physical devices (or things) in the physical world and is sent to Communication Networks and the Information domain for further processing. Devices may communicate with each other either via the communication network (with or without a gateway) or directly without using the communication network or combinations of these communication links. The exchange of information not only happens between physical things in the Physical world but also between virtual things in the Information World.
The communication networks provide capabilities for reliable and efficient data transfer. The network infrastructure may be implemented or realized via existing networking technologies (e.g., TCP-IP networks) or evolving networks following the current telecommunication trends.
Other Applications
Beyond these sectors, AI is used across various industries:
Retail: Personalizing shopping experiences and managing inventory.
Transportation: Developing self-driving cars and improving traffic management.
Energy: Enhancing energy efficiency and predicting demand.
Government: Improving public safety, crime detection, and citizen services.
IoT Challenges
Scalability: Many contemporary IoT Applications and Systems include very large numbers of connected devices. As these networks grow, device management and coordination become increasingly challenging, as in many cases the rapid growth of connected nodes or the increased number of data flow might require substantial infrastructure changes.
Solution: One solution could be to use scalable architectures like edge computing, and distributed processing, and employ load balancing for the efficient handling of numerous devices.
Network Congestion: This is also related to the scalability problem mentioned above since the increased number of connected devices might cause traffic/congestion in the network which will degrade the quality of service given the increase in the packet loss, the associated delays and other issues.
Solution: Possible solutions include the optimization of communication protocols, the use data compression and prioritization of critical data
Security: IoT devices are typically operated on low-power, low-processing capability electronics which does not allow for security mechanisms to be efficiently implemented on them. Given the significant increase in firmware vulnerabilities, IoT devices usually make the perfect back door to enter a secure network.
Solution: Implement encryption, authentication, access controls, and regular updates. Use intrusion detection and anomaly detection for early identification of security threats.
Device Management. Managing numerous IoT devices becomes large especially if they are heterogeneous, and come with many complicated features, authentication mechanisms, update requirements etc.
Solution: Employ device management platforms for automated tasks like updates and monitoring. Implement standardized protocols such as MQTT and CoAP.
Interoperability:Typically, in large IoT systems, the various components (sensors, actuators, microcontrollers etc) might come from different vendor and given the not so standardized IoT framework they may create interoperability issues.. Adjustments may be needed when adding new hardware and software to maintain functionality and accommodate innovative technology.
Solution: A possible solution is to Adopt industry standards for communication and data formats. Use middleware solutions to handle different protocols.
Power Consumption: There exist many IoT applications that are installed in remote places or in spaces where providing power to them could be very challenging (e.g. at the bottom of the lake to monitor pollution). This means that either we need have batteries that last for a long time or to limit power consumption but ideally both.
Solution: Optimize communication protocols, use low-power technologies like LPWAN, and design energy-efficient hardware or push some functionality to a central processing unit.
Data Privacy: A big concern in IoT Systems is about what happens with the collected data especially if they are sensitive ones (e.g. heath data in an IoT e-health system).
Solution: Use encryption, data anonymization, and explicit privacy policies in practice. Respect laws like the GDPR and HIPAA.
Healthcare
AI aids in disease diagnosis, treatment development, and personalized care:
Disease Diagnosis: Analysing patient data for early and accurate disease detection.
Treatment Development: Using data to develop new drugs and therapies.
Personalized Care: Tailoring treatment plans based on individual patient data.
Remote Sensing Advantages
As can be seen from the historical review, remote sensing has significantly improved in terms of recording data related to space, time and radiation. This means that the Earth's surface is frequently recorded at higher spatial resolutions and more parts of the electromagnetic spectrum. Remote sensing is considered a modern, specialized tool that finds applications in many scientific subjects, including environmental sciences, forestry, geology, archaeology, oceanography, etc. Among the advantages that justify the use of remote sensing for acquiring and processing data from our planet, the following are considered very significant.
Summary Coverage: A satellite image covers a large area, requiring hundreds of aerial photographs and even more thousands of hours of sample collection.
Repeated Coverage: The frequency of recordings is very important, especially in analyses of various series. The frequency, similar to the satellite system, varies from a few hours to several days.
Accessibility: In remote or inaccessible areas, such as examples of desert areas, oceans, tropical forests, etc., remote sensing provides the possibility of collecting and analyzing data.
Data Homogeneity: Information and data recorded by satellite systems provide uniform data regarding the spatial and visual performance of the recording objects and data.
Multispectral Data Characteristics: Data recorded in different parts of the electromagnetic spectrum provide more possibilities for the display of the desired information.
Digital Data Form: It offers the possibility of digital analysis using specialized software in all the advantages that this entails.
Recording Time Duration: The required recording time is very short, which means a minimum change in spatial and visual alterations that may arise from environmental changes.
Data Cost: The relatively low cost of recording data combined with the capabilities they offer is also a significant advantage.
Remote Sensing Disadvantages
Although significant progress has been made since the advent of remote sensing, some problems related to its application in environmental issues remain unsolved. One of the most serious problems concerns the recorded data, which contain errors due to atmospheric conditions, the topography of the area and the functioning of the satellite system. These errors are related to differences between the actual and the satellite-recorded reflected radiation. It is very important for the further successful application of these data to adapt the data as closely as possible to the actual values, especially in cases where the analysis concerns temporal studies.
Another problem that scientists face is the difficulty in understanding the nature and mechanism of even very simple relationships and interactions between the radiation recorded by the satellite and the target objects. The changes that occur in the conditions of the atmosphere, lithosphere and hydrosphere are so large in spatial and temporal dimensions and the mechanisms of interaction between energy and matter are so complex, that simple object-detector relationships are difficult to determine.
Finance:
AI applications in finance include personalized services, risk management, and operational automation:
Risk and Fraud Detection: Identifying potentially fraudulent activities swiftly.
Personalized Recommendations: Offering tailored financial advice and services.
Document Processing: Extracting and analysing data from documents for tasks like loan processing.
IoT Requirements
Identification-based connectivity: There needs to be a support for the “Things” to be connected to the IoT based on their identifiers (IDs) which might be heterogeneous hence some unified processing is required.
Interoperability: Interoperability between heterogeneous and distributed systems needs to be ensured.
Automatic Networking: The IoT network infrastructure should provide control functions for automatic networking including self-management, self-configuration, self-healing, self-optimization and self-protection, to be able to support and facilitate adaptation in different application domains, different communication environments and larger numbers and types of devices.
Autonomic services provisioning: Services need to be provided by automatically capturing, communicating and processing of the data of the “Things” according to the rules configured by the operators and/or configured by the subscribers.
Location-based capabilities: Localization is a key enabling technology in IoT as location-based services must be supported. Things should be able to track their position to facilitate the provision of services which depend on their location.
Security: The ability of any Thing to connect at any time and any place generates significant security threats against CIA (Confidentiality, Integrity and Authenticity) for both data and services. Therefore, there is an important requirement to integrate different security policy and measures related to the things and their communication in an IoT framework.
Privacy protection: Data acquired by “Things” may contain private information of their owners and/or their users. Therefore, privacy protection must be supported during transmission, aggregation, storage, mining and processing of this data while not setting a barrier to data source authentication.
High quality and highly secure human body related services: Services which are based on the capturing, communicating, and processing of data related to human behaviour (e.g. exercise, health, location etc.) automatically or through human intervention should be offered while guaranteeing high quality, accuracy and security.
Plug and Play: It is important for IoT systems to support plug and play capability in order to enable or facilitate on-the-fly generation, composition and acquisition of semantic-based configurations to seamlessly integrate an internetwork of things with the respective applications and efficiently respond to these applications’ requirements.
Manageability: Applications in an IoT systems usually need to work automatically without the intervention or participation of people and therefore the whole operation process needs to be manageable by the relevant entities in order to ensure normal network operations.
Introduction to Deep Technologies
citizensinpower
Created on February 17, 2025
Start designing with a free template
Discover more than 1500 professional designs like these:
View
Customer Service Course
View
Dynamic Visual Course
View
Dynamic Learning Course
View
Akihabara Course
Explore all templates
Transcript
Introduction to Deep Technologies
PREVENT Project
Start
Definition and scope of deep technologies
"Deep technology" encompasses advanced technologies that tackle complex global issues such as climate change and healthcare with innovative solutions. Unlike traditional technologies, deep tech focuses on addressing local problems proactively and efficiently. Key areas of deep tech include Artificial Intelligence, Space Tech, and Computational Biology, among others. These technologies are distinguished by their disruptive potential and reliance on cutting-edge scientific breakthroughs.
Index
Definition of Deep Technologies
Remote Sensing
Robotics
Artificial Intelligence
Computer Vision
IoT and communications
Objectives
Through this module, learners will gain the role of advanced technologies in disaster prevention and mitigation. By exploring real-world case studies, scientific research, and interactive simulations, students will develop the ability to classify different types of natural disasters, analyze their social, economic, and environmental consequences, and evaluate the impact of climate change on disaster frequency and severity. Furthermore, learners will acquire critical thinking skills to assess disaster risk management strategies, understand how deep technologies such as AI, IoT, and satellite imaging contribute to early warning systems, and explore innovative disaster resilience frameworks.
By the end of the module, students will be able to apply knowledge in risk assessment, disaster response planning, and climate adaptation strategies, equipping them with essential skills for careers in environmental science, emergency management, and sustainable development.
'True learning begins when knowledge inspires action, and understanding drives change. Equip yourself to turn challenges into opportunities.'
'Harnessing the power of new technologies, we can predict, prepare for, and mitigate the impact of natural disasters—turning technology into a lifeline for our future.'
01
Definition of Deep Technologies
01
Definition of Deep Technologies
Deep Technologies refer to cutting-edge innovations that are based on substantial scientific and engineering advancements. These technologies often require significant research and development (R&D) and have the potential to create disruptive, transformative changes across various industries. Unlike incremental technological improvements, deep tech innovations are rooted in fundamental scientific breakthroughs and complex engineering. Key Characteristics of Natural Disasters:
Deep Tech in the Context of Natural Disasters
When combined with natural disaster preparedness, response, and mitigation, deep technologies can offer groundbreaking solutions. Examples include:
02
Artificial Intelligence
02
Introduction to AI
Artificial Intelligence (AI) is a branch of computer science focused on creating systems that can perform tasks traditionally requiring human intelligence. These tasks include learning, reasoning, problem-solving, perception, language understanding, and decision-making. AI systems can analyze vast amounts of data, recognize patterns, and improve their performance over time through machine learning algorithms.
AI is classified into different categories based on capabilities and functions.
The primary types include:
- Narrow AI (Weak AI): AI systems designed to perform a specific task, such as facial recognition, speech translation, or medical diagnosis. These are the most commonly used AI applications today.
- General AI (Strong AI): A theoretical AI system capable of understanding, learning, and performing any intellectual task that a human can do. This level of AI is still under research and development.
- Super AI: A hypothetical future stage where AI surpasses human intelligence in all aspects, including creativity, problem-solving, and decision-making.
AI technologies rely on various subfields, including:AI applications
Generic Applications
Business Intelligence (BI)
Observation systems
Healthcare
Education
Manufacturing
Finance
Other Applications
03
Remote Sensing
03
Introduction to Remote Sensing
Remote sensing describes the collection of data about an object, area, or phenomenon from a distance with a device that is not in contact with the object”. Remote sensing is generally broadly defined as the acquisition of data and information about Earth's surfaces and objects without physical contact. Its exact definition is challenging and definitions like the one above is often considered too general. There are many ways to define remote sensing in the literature. Two common elements characterize this scientific field a) the concept of "information collection" and b) "observation from a distance"
Types and causes of Floods
Remote sensing has gone through many stages to reach its current level. It is a quite evolutionary scientific field that relies on various other scientific disciplines such as mathematics, physics, computer science, etc. Since the discovery of photography in the 19th century, significant progress has been made in environmental remote sensing. Although it is quite difficult to precisely define the starting point of remote sensing and its evolutionary phases, five main stages are distinguished during its development . While Remote Sensing (RS) is its own field, it often acts as a complement to GIS analyses, adding unique information and analysis techniques. There are two types of RS, active and passive and they are generally used for different applications. Active RS involves sending out a signal and waiting for its return to the sensor. RADAR and LIDAR are examples of active RS, as they send out energy, microwave and laser pulses respectively and record the signals as they bounce back.
Remote Sensing Disadvantages
Remote Sensing Advantages
03
Remote Sensing for Natural Disaster Monitoring and Mitigation
Soil moisture is a critical variable for understanding water resources and natural hazards such as floods and landslides. Accurate estimation of spatial and temporal variations in moisture is important for environmental studies and improving flood forecasts, especially in medium and large drainage basins where floods often cause disasters. The condition of surface soil moisture is critical for assessing rainwater infiltration or runoff, so this information is vital for flood prediction models. Moreover, soil moisture in mountainous areas is a key factor for landslides. Since the 1970s, research in this field has used techniques that cover the entire electromagnetic spectrum, from optical to microwave. These techniques vary in terms of wavelength range, source of electromagnetic energy, sensor response and the physical relationship between response and soil moisture. The use of solar radiation measures the reflected sunlight from the Earth's surface. However, microwaves and thermal infrared have been used more frequently for calculating soil moisture. The use of the thermal section is based on measuring the surface temperature of the soil and methods such as thermal inertia and the temperature/vegetation index. Thermal inertia is simple and accurate in areas with minimal or no vegetation. In summary, remote sensing through active and passive sensors provides valuable data for estimating soil moisture, improving natural hazard prediction and water resource management.
04
Robotics
04
Introduction to Robotics
Robotics, a multidisciplinary field at the intersection of engineering, computer science, and artificial intelligence, plays an increasingly vital role in addressing the complex challenges of natural disaster management. Robots are designed to perform tasks that are dangerous, repetitive, or otherwise unsuitable for humans, making them invaluable in disaster scenarios. The evolution of robotics, driven by advancements in sensors, machine learning, and autonomous systems, has expanded their capabilities and applications, enabling more effective and efficient responses to natural disasters. In disaster management, robotics can perform a variety of crucial functions, from search and rescue operations to environmental monitoring and infrastructure inspection. Autonomous drones, for instance, can quickly survey affected areas, providing real-time data and high-resolution imagery to aid in decision-making and coordination efforts. Ground robots can navigate hazardous terrains to locate survivors, deliver supplies, and perform structural assessments. Furthermore, underwater robots are essential for operations in flooded or submerged environments.
Different categories of robotics in disaster management Search and Rescue Robots – Assist in locating and saving survivors in collapsed structures or hazardous zones. Firefighting Robots – Suppress wildfires and prevent fire spread using AI and automated water or foam dispersal. Aerial Drones – Provide real-time surveillance, assess damage, and identify hazards from above. Underwater and Amphibious Robots – Aid in flood rescues and assess underwater damage. Inspection and Recovery Robots – Evaluate structural integrity and assist in clearing debris post-disaster.
Robotics aids disaster response by enhancing rescue, fire suppression, and damage assessment with AI and sensors.
04
History of Robotics
Robotics has significantly advanced disaster response, from early industrial automation to AI-powered autonomous systems. Robots assist in search and rescue, hazard assessment, and crisis recovery. Key deployments include Chernobyl (1986), 9/11 (2001), and Fukushima (2011), where robots navigated hazardous environments to support emergency teams. Today, AI-driven drones, autonomous vehicles, and robotic systems enhance real-time data collection, structural analysis, and aid delivery. As technology evolves, robotics continues to improve disaster management, making responses faster, safer, and more effective.
05
Iot and Communications
05
Introduction to IoT and communications
Internet of Things has become a very popular topic of research and Innovation mainly due to the ubiquitous transformation of computing. Physical devices have become “smart”, being able to sense, communicate in a pervasive way and interact with their environment offering useful applications and solutions to humankind. Today they find applications in a wide range of activities like Health, Transportation, Agriculture, Home and Industrial Automation, Retail and many more. It is expected an exponential growth of network connections which should be facilitated by powerful networks.
Fundamental Characteristics of the IoT and Communications
Interconnectivity: Any IoT device can be interconnected with the global Information and Communication Infrastructure. Things-related services: The IoT provisions services which concern the connected “things” within their constraints such as privacy protections and semantic consistency between physical things and their associated virtual things. Heterogeneity: Heterogeneous IoT devices with different hardware and networking characteristics get connected and interact with other devices or platforms on various types networks. Dynamic Changes: While roaming and interacting in an IoT system, devices change their state dynamically. For example, sleeping and waking up, get connected or disconnected while changing their location and speed. Additionally and equally important the number of connected devices changes dynamically. Enormous scale: Usually the number of devices that need to be managed and that of the devices that communicate with each other is significantly larger than the ones that are connected to the Internet. This practically means that the underlying communication network fabric needs to be able to support the high volume of data and the rate and quality that this needs to be exchanged.
Aspects of IoT
Useful Definitions
IoT Architecture
Observation systems
Fundamental Characteristics
IoT Requirements
IoT Devices and Components
IoT Challenges
IoT Challenges
Natural Disaster IoT Applications
Real-Time Data Collection Collecting real-time data is crucial for disaster prediction, detection, and management. Sensors measuring air quality, humidity, and temperature help authorities assess risks and take preventive measures. Interconnected Devices IoT systems integrate diverse devices with different capabilities for resource allocation and strategic response in disasters. Early Warnings & Predictions AI and machine learning analyze real-time data to predict disasters, enabling early intervention and saving lives. Data Sharing & Analysis Effective disaster management requires collaboration among first responders, authorities, and the public through data sharing. Human-less Interaction Drones and robots aid disaster response by performing remote monitoring, search, and intervention in hazardous areas. Smart Infrastructure & Resilience IoT-enabled buildings with sensor networks enhance resilience by monitoring structures, shutting down utilities, and activating fire suppression systems to reduce disaster impact.
06
Computer Vision
06
Computer Vision
Computer vision enables machines to interpret and analyze visual data, automating tasks like object detection, image segmentation, and 3D scene reconstruction. Advances in AI, computational power, and deep learning, particularly CNNs, have revolutionized the field, making it integral to applications such as autonomous vehicles, facial recognition, and augmented reality. In disaster management, computer vision enhances response teams by enabling real-time monitoring, search and rescue, and damage assessment. By providing accurate visual data, it aids decision-making during emergencies. This section aims to help educators understand computer vision, its role in disaster response, and how to integrate it into education, preparing students to use these technologies for crisis preparedness.
06
History of Computer Vision
Computer vision emerged in the 1960s-70s, focusing on 2D image analysis and basic shape recognition. The 1980s introduced motion detection, optical flow, and industrial applications. In the 1990s, machine learning techniques like SVM and PCA improved face recognition and object detection. The 2000s saw the rise of CNNs, revolutionizing image analysis with deep learning, highlighted by AlexNet (2012). The 2010s brought advanced architectures (VGG, ResNet) and real-world applications like autonomous vehicles and medical imaging. In the 2020s, innovations like GANs, transformers, and self-supervised learning push the field forward. Generative AI enables image synthesis, data augmentation, and AI-driven simulations, enhancing training and real-world applications. Computer vision now plays a crucial role in disaster management, healthcare, and environmental monitoring, solving global challenges.
Thank you!
Useful Definitions
Device: In the IoT context, this is a piece of equipment must be able to communicate and could optionally sense, act, capture data, store data or process data. Its only mandatory capability is the communication. Thing: An object inside the IoT system which is capable of being identified and integrated into communication system. Physical Thing: An object of the physical world which is able of being sensed, actuated and connected is known as a physical thing (e.g. industrial robots, electrical equipment etc.) Virtual Thing: An object in the information world capable of being stored, processed and accessed is known as virtual thing. For example, multimedia content, application software etc. Internet of Things: A global information infrastructure which enables advanced services by interconnecting Things (Physical and/or virtual) based on existing and/or evolving interoperable technologies. The IoT includes functions for identification, data capture, processing, and communication to offer different kinds of applications whilst ensuring security and privacy.
Fundamental Characteristics
Interconnectivity: Any IoT device can be interconnected with the global Information and Communication Infrastructure. Things-related services: The IoT provisions services which concern the connected “things” within their constraints such as privacy protections and semantic consistency between physical things and their associated virtual things. Heterogeneity: Heterogeneous IoT devices with different hardware and networking characteristics get connected and interact with other devices or platforms on various types networks. Dynamic Changes: While roaming and interacting in an IoT system, devices change their state dynamically. For example, sleeping and waking up, get connected or disconnected while changing their location and speed. Additionally and equally important the number of connected devices changes dynamically. Enormous scale: Usually the number of devices that need to be managed and that of the devices that communicate with each other is significantly larger than the ones that are connected to the Internet. This practically means that the underlying communication network fabric needs to be able to support the high volume of data and the rate and quality that this needs to be exchanged.
Business Intelligence (BI)
AI-powered BI tools enhance data collection, analysis, and visualization, improving decision-making and productivity while reducing costs. Data Collection: Gathering structured (e.g., databases) and unstructured data (e.g., text, images). Data Analysis: Identifying patterns, trends, and relationships in data. Data Visualization: Creating visual representations for easier comprehension. Decision-Making: Providing insights and recommendations to support data-driven decisions.
Education
AI can personalize learning, enhance student engagement, and automate administrative tasks: Personalized Learning: Adapting learning experiences based on student progress. Improved Engagement: Providing interactive learning experiences and real-time feedback. Administrative Tasks: Automating tasks like grading and scheduling to free up teacher time.
Generic Applications
Natural Language Processing (NLP): NLP enables computers to understand and generate human language. Applications include machine translation, spam filtering, and sentiment analysis. Computer Vision: Computer vision allows computers to interpret visual content, used in applications like self-driving cars, facial recognition, and object detection. Machine Learning (ML): ML enables computers to learn from data and improve performance over time. Applications include predictive analytics, fraud detection, and recommendation systems. Robotics: Robotics involves designing, constructing, and operating robots for applications such as manufacturing, healthcare, and space exploration.
Manufacturing
AI improves manufacturing efficiency, productivity, and quality control: Improved Efficiency: Automating tasks such as assembly and inspection. Increased Productivity: Optimizing production processes. Quality Improvement: Detecting defects and enhancing quality control.
IoT Devices and Components
Regarding the processing capabilities devices are classified as: Devices with no processing capability: Passive devices, usually low-cost and with no microcontrollers. A typical example is an RFID. Devices with low processing capabilities: Their processing capabilities are limited to the reading and writing data from or to sensors and actuators and sending this data to IoT applications, but they are not able to make decisions or run complex algorithm. They are typically low cost and usually embed a very low-power and low-cost microcontroller. A typical example is a smart light or a door sensor. Devices with high procession capabilities: They have enough processing power to enable them to make decisions and run complex algorithms. They are typically high cost as they employ a powerful microcontroller. (e.g. a smart cooling system, or a smart thermostat) Regarding the connectivity capabilities devices can be classified as: Devices with low connectivity: This kind of devices do not connect directly to the communication network to transfer the data but instead they rely on additional elements (e.g. gateway) to perform communications tasks (e.g. protocol translation or internet connectivity). Devices with High connectivity: They have the hardware and ability to directly connect to the network to transfer the data.
IoT Architecture
A physical thing can be mapped (or represented) by one or more virtual things in the Information domain. Information is being collected by physical devices (or things) in the physical world and is sent to Communication Networks and the Information domain for further processing. Devices may communicate with each other either via the communication network (with or without a gateway) or directly without using the communication network or combinations of these communication links. The exchange of information not only happens between physical things in the Physical world but also between virtual things in the Information World. The communication networks provide capabilities for reliable and efficient data transfer. The network infrastructure may be implemented or realized via existing networking technologies (e.g., TCP-IP networks) or evolving networks following the current telecommunication trends.
Other Applications
Beyond these sectors, AI is used across various industries: Retail: Personalizing shopping experiences and managing inventory. Transportation: Developing self-driving cars and improving traffic management. Energy: Enhancing energy efficiency and predicting demand. Government: Improving public safety, crime detection, and citizen services.
IoT Challenges
Scalability: Many contemporary IoT Applications and Systems include very large numbers of connected devices. As these networks grow, device management and coordination become increasingly challenging, as in many cases the rapid growth of connected nodes or the increased number of data flow might require substantial infrastructure changes. Solution: One solution could be to use scalable architectures like edge computing, and distributed processing, and employ load balancing for the efficient handling of numerous devices. Network Congestion: This is also related to the scalability problem mentioned above since the increased number of connected devices might cause traffic/congestion in the network which will degrade the quality of service given the increase in the packet loss, the associated delays and other issues. Solution: Possible solutions include the optimization of communication protocols, the use data compression and prioritization of critical data Security: IoT devices are typically operated on low-power, low-processing capability electronics which does not allow for security mechanisms to be efficiently implemented on them. Given the significant increase in firmware vulnerabilities, IoT devices usually make the perfect back door to enter a secure network. Solution: Implement encryption, authentication, access controls, and regular updates. Use intrusion detection and anomaly detection for early identification of security threats. Device Management. Managing numerous IoT devices becomes large especially if they are heterogeneous, and come with many complicated features, authentication mechanisms, update requirements etc. Solution: Employ device management platforms for automated tasks like updates and monitoring. Implement standardized protocols such as MQTT and CoAP. Interoperability:Typically, in large IoT systems, the various components (sensors, actuators, microcontrollers etc) might come from different vendor and given the not so standardized IoT framework they may create interoperability issues.. Adjustments may be needed when adding new hardware and software to maintain functionality and accommodate innovative technology. Solution: A possible solution is to Adopt industry standards for communication and data formats. Use middleware solutions to handle different protocols. Power Consumption: There exist many IoT applications that are installed in remote places or in spaces where providing power to them could be very challenging (e.g. at the bottom of the lake to monitor pollution). This means that either we need have batteries that last for a long time or to limit power consumption but ideally both. Solution: Optimize communication protocols, use low-power technologies like LPWAN, and design energy-efficient hardware or push some functionality to a central processing unit. Data Privacy: A big concern in IoT Systems is about what happens with the collected data especially if they are sensitive ones (e.g. heath data in an IoT e-health system). Solution: Use encryption, data anonymization, and explicit privacy policies in practice. Respect laws like the GDPR and HIPAA.
Healthcare
AI aids in disease diagnosis, treatment development, and personalized care: Disease Diagnosis: Analysing patient data for early and accurate disease detection. Treatment Development: Using data to develop new drugs and therapies. Personalized Care: Tailoring treatment plans based on individual patient data.
Remote Sensing Advantages
As can be seen from the historical review, remote sensing has significantly improved in terms of recording data related to space, time and radiation. This means that the Earth's surface is frequently recorded at higher spatial resolutions and more parts of the electromagnetic spectrum. Remote sensing is considered a modern, specialized tool that finds applications in many scientific subjects, including environmental sciences, forestry, geology, archaeology, oceanography, etc. Among the advantages that justify the use of remote sensing for acquiring and processing data from our planet, the following are considered very significant. Summary Coverage: A satellite image covers a large area, requiring hundreds of aerial photographs and even more thousands of hours of sample collection. Repeated Coverage: The frequency of recordings is very important, especially in analyses of various series. The frequency, similar to the satellite system, varies from a few hours to several days. Accessibility: In remote or inaccessible areas, such as examples of desert areas, oceans, tropical forests, etc., remote sensing provides the possibility of collecting and analyzing data. Data Homogeneity: Information and data recorded by satellite systems provide uniform data regarding the spatial and visual performance of the recording objects and data. Multispectral Data Characteristics: Data recorded in different parts of the electromagnetic spectrum provide more possibilities for the display of the desired information. Digital Data Form: It offers the possibility of digital analysis using specialized software in all the advantages that this entails. Recording Time Duration: The required recording time is very short, which means a minimum change in spatial and visual alterations that may arise from environmental changes. Data Cost: The relatively low cost of recording data combined with the capabilities they offer is also a significant advantage.
Remote Sensing Disadvantages
Although significant progress has been made since the advent of remote sensing, some problems related to its application in environmental issues remain unsolved. One of the most serious problems concerns the recorded data, which contain errors due to atmospheric conditions, the topography of the area and the functioning of the satellite system. These errors are related to differences between the actual and the satellite-recorded reflected radiation. It is very important for the further successful application of these data to adapt the data as closely as possible to the actual values, especially in cases where the analysis concerns temporal studies. Another problem that scientists face is the difficulty in understanding the nature and mechanism of even very simple relationships and interactions between the radiation recorded by the satellite and the target objects. The changes that occur in the conditions of the atmosphere, lithosphere and hydrosphere are so large in spatial and temporal dimensions and the mechanisms of interaction between energy and matter are so complex, that simple object-detector relationships are difficult to determine.
Finance:
AI applications in finance include personalized services, risk management, and operational automation: Risk and Fraud Detection: Identifying potentially fraudulent activities swiftly. Personalized Recommendations: Offering tailored financial advice and services. Document Processing: Extracting and analysing data from documents for tasks like loan processing.
IoT Requirements
Identification-based connectivity: There needs to be a support for the “Things” to be connected to the IoT based on their identifiers (IDs) which might be heterogeneous hence some unified processing is required. Interoperability: Interoperability between heterogeneous and distributed systems needs to be ensured. Automatic Networking: The IoT network infrastructure should provide control functions for automatic networking including self-management, self-configuration, self-healing, self-optimization and self-protection, to be able to support and facilitate adaptation in different application domains, different communication environments and larger numbers and types of devices. Autonomic services provisioning: Services need to be provided by automatically capturing, communicating and processing of the data of the “Things” according to the rules configured by the operators and/or configured by the subscribers. Location-based capabilities: Localization is a key enabling technology in IoT as location-based services must be supported. Things should be able to track their position to facilitate the provision of services which depend on their location. Security: The ability of any Thing to connect at any time and any place generates significant security threats against CIA (Confidentiality, Integrity and Authenticity) for both data and services. Therefore, there is an important requirement to integrate different security policy and measures related to the things and their communication in an IoT framework. Privacy protection: Data acquired by “Things” may contain private information of their owners and/or their users. Therefore, privacy protection must be supported during transmission, aggregation, storage, mining and processing of this data while not setting a barrier to data source authentication. High quality and highly secure human body related services: Services which are based on the capturing, communicating, and processing of data related to human behaviour (e.g. exercise, health, location etc.) automatically or through human intervention should be offered while guaranteeing high quality, accuracy and security. Plug and Play: It is important for IoT systems to support plug and play capability in order to enable or facilitate on-the-fly generation, composition and acquisition of semantic-based configurations to seamlessly integrate an internetwork of things with the respective applications and efficiently respond to these applications’ requirements. Manageability: Applications in an IoT systems usually need to work automatically without the intervention or participation of people and therefore the whole operation process needs to be manageable by the relevant entities in order to ensure normal network operations.