1) AI Threat detection
Threat detection using AI is crucial in modern cybersecurity because of the increasing complexity and scale of cyber threats. Traditional rule-based security systems struggle to keep up with evolving attacks, such as zero-day exploits and advanced persistent threats (APTs). AI models trained on real-world traffic data can automatically detect suspicious patterns and anomalies in network activity, reducing response time and improving defense mechanisms. In practice, companies use AI-driven Security Information and Event Management (SIEM) systems to detect intrusions in enterprise networks. The challenge of creating an AI-powered threat detection system mirrors the work of cybersecurity analysts who must filter through massive amounts of data to identify genuine threats while minimizing false positives.
In this challenge, you must develop an AI model that analyzes network or system logs to identify potential security threats. You can use the UNSW-NB15 dataset or the CICIDS2017 dataset to learn to detect attacks.
2) Duckietown Autonomous Navigation Challenge
Program one or several Duckietown robots to follow a course and obey traffic rules. The robots must follow traffic rules, detect obstacles, and make intelligent decisions in real time. Teams can use machine learning, reinforcement learning, or classical computer vision techniques to improve navigation accuracy and efficiency.
Autonomous navigation is a fundamental challenge in self-driving cars, drones, and industrial automation. AI-powered perception and decision-making are critical for ensuring safe and efficient transportation in real-world environments. Self-driving technology is already being deployed in industries like logistics, agriculture, and urban mobility, where vehicles must adapt to dynDuckietown Autonomous Navigation ChallengeDuckietown Autonomous Navigation Challengeamic surroundings without human intervention. The Duckietown challenge provides a hands-on introduction to these challenges on a smaller scale, allowing participants to experiment with obstacle detection, lane following, and real-time AI decision-making—key problems faced by engineers working on autonomous vehicles today.
3) Secure the Smart Home
Smart home devices—such as security cameras, smart thermostats, voice assistants, and connected door locks—offer convenience but also introduce serious security risks. Hackers can exploit vulnerabilities to spy on users, steal data, or even take control of devices. Many IoT devices have weak security configurations, making them prime targets for cyberattacks like botnet infections (e.g., Mirai botnet). Securing IoT networks is a critical challenge for both consumers and manufacturers. This challenge mirrors real-world efforts by cybersecurity professionals to develop secure-by-design IoT systems and proactive threat detection mechanisms.
Teams in this challenge may work on intrusion detection for smart home networks, securing IoT device communication, or detecting vulnerabilities in connected devices. Solutions can range from AI-powered anomaly detection systems to encryption methods that protect user data. You can use datasets like the TON_IoT Dataset or the Bot-IoT Dataset.
4) AI for Misinformation Detection
Misinformation spreads rapidly on social media, influencing public opinion and even political outcomes. AI-powered fact-checking tools are essential for combating fake news, ensuring reliable information sources, and enhancing media literacy.
Develop an AI system that can detect fake news, misleading content, or deepfake videos/images. Teams can focus on text-based misinformation, image manipulation detection, or social media trend analysis.
5) AI for Climate and Environmental Monitoring
Governments, researchers, and environmental organizations are increasinlgy seeking AI solutions to respond to disasters, optimize resource usage, and develop sustainable policies. In this challenge, you must analyze satellite images, sensor data, or weather patterns to monitor climate change, predict natural disasters, or detect environmental pollution.
Some possible applications are deforestation tracking, air quality prediction, or wildfire detection. Some datasets that can be useful are the NASA Landsat Satellite Imagery, Global Forest Change Dataset, the Berkeley Earth Surface Temperature Data, or the NOAA Global Historical Tsunami Database.
6) Smart Agriculture Monitoring System with AI
Design a comprehensive smart agriculture monitoring system that leverages AI to optimize resource usage and crop yields. The system should collect data from various IoT sensors deployed across fields (soil moisture, temperature, light levels, pest detection), process this data at the edge, and provide real-time insights and autonomous decisions without constant cloud connectivity. The solution should address precision agriculture needs, including irrigation optimization, pest management, and harvest timing, while considering the constraints of rural deployments such as limited connectivity and power sources.
7) Adapting CNNs for Image Processing on IoT Devices
Convolutional Neural Networks (CNNs) were successfully used for object detection tasks. Some of these networks can fit into tiny devices (e.g., ESP32-EYE, ESP32-CAM, ESP32S3 WROOM). One example of CNN application in the agricultural domain is the flowmeter used to measure water for irrigation: the goal is to automatically identify numbers from images of flowmeters. The following image is an example
The challenge is to develop a solution for automatic number recognition from images with computer vision techniques. However, there is an important restriction: the solution should fit into limited memory (e.g., 4 MB), so that it can be further deployed into an ESP32 device.
Using an analytical approach, develop a use case and identify the process you would follow to solve the problem where CNNs are adapted to tiny IoT devices.
In case someone wants to try solving the specific challenge on flowmeter detection described above, we can also provide images (about 100) and annotation files (in JSON format).
8) Self-optimizing production test system using AI and IoT technologiesXX
At the core of Industry 4.0, the integration of technologies such as the Internet of Things (IoT) and Artificial Intelligence (AI) is radically transforming manufacturing. This challenge focuses on developing and idea for a production test system that is not only intelligent but also autonomous, capable of self-adjusting and optimizing in real-time. The key lies in AI’s ability to interpret complex data from IoT sensors scattered throughout the plant and use these insights to make instant decisions that improve operational efficiency, product quality, and environmental sustainability.
We seek solutions that can autonomously monitor, predict, and adapt manufacturing and test processes in real-time, reducing downtime and increasing efficiency. The goal is to create a smart manufacturing environment where machines can predict failures before they happen, adjust production parameters to optimize energy use and material efficiency, and ensure the highest quality of the final product.
The challenge aims to develop a self-optimizing production test system using AI and IoT technologies within Industry 4.0. It seeks to create a system that can analyse data from smart sensors to predict and prevent failures, optimize production in real-time, automate product quality assurance, and enhance energy efficiency and sustainability. Innovators, engineers, developers, and creative thinkers are invited to collaborate in building a smarter, more efficient, and sustainable manufacturing future.
9) “IoT Lab Hub” – a hybrid learning laboratory for IoT
The Internet of Things (IoT) revolution is here, bringing with it the need to educate and inspire the next generation of engineers, designers, and thinkers in IoT. We present an exciting challenge: conceptualize the “IoT Lab Hub,” an innovative educational laboratory that combines both physical and virtual hands-on experiences, aimed at MERIT master’s students interested in exploring the vast world of IoT.
This hybrid lab combines in-situ physical experimentation with virtual simulation capabilities, providing students an unparalleled opportunity to delve into the Internet of Things (IoT) from anywhere. This dual approach ensures that learning and innovation are not confined by geographical boundaries. The challenge encompasses the development of sophisticated simulation tools to model complex IoT networks and scenarios. It should create a collaborative platform for students, educators, and industry experts to share knowledge and mentorship. A comprehensive digital repository of IoT resources, including practical tutorials and cutting-edge case studies, is also part of the vision.
The challenge invites individuals to contribute their innovative ideas and passion for shaping the future of IoT education. It offers an opportunity to impact how future generations learn and apply IoT concepts. Join in designing the “IoT Lab Hub” to become a pivotal resource for IoT education.
10) The Convergence of Artificial Intelligence and Computer Vision in Optimizing Production Efficiency
The efficiency in industrial production is an ongoing challenge. Companies strive to maximize productivity, minimize downtime, and reduce defects on the production line. However, identifying areas for improvement and making data-driven decisions can be complex. Calculating OEE (Overall Equipment Efficiency) is crucial for assessing machinery and processing performance, but traditionally, it has been a manual and error-prone process.
This is where artificial intelligence (AI) and computer vision come into play. By analyzing real-time data and detecting patterns, AI tools can identify inefficiencies and predict potential issues before they occur. Computer vision, on the other hand, allows for visual inspection of products and precise defect detection. By combining these technologies, companies can optimize production, reduce downtime, and enhance product quality.
11) Harnessing Blockchain for design projects
Imagine a scenario where the integrity of design projects is paramount, from the initial concept to the final production stage. How can you ensure the chain of custody of images, copyrights, additional documentation, and notes throughout the design process, validation of the final proposal by the client, and subsequent production (printing of series of the final document), guaranteeing that nothing has changed since it was validated until it is produced? This is the intricate puzzle we invite you to solve.
Your mission is to incorporate Blockchain and NFT technologies into the workflow of graphic design projects to safeguard their authenticity, integrity, and ownership. From the moment an idea is conceived to its realization in the form of a tangible product, your solution should ensure that every step of the process is transparent, immutable, and tamper-proof.
This challenge not only calls for expertise in areas such as Artificial Intelligence, and Cybersecurity, but also demands a deep understanding of the creative process and the complexities of intellectual property rights. We’re seeking teams that can blend technical prowess with creative flair to devise innovative solutions that redefine the standards of trust and security in the realm of design.
With your ingenuity and expertise, you have the opportunity to shape the future of creative ventures, ensuring that every stroke of genius is protected and every masterpiece is authenticated. Join us in this exhilarating challenge and be at the forefront of a paradigm shift in the world of design. Get ready to push the boundaries of innovation and leave an indelible mark on the industry.
12) Revolutionizing communication in remote environments
Welcome to our exhilarating competition, where we invite you to explore the frontiers of technology and innovation. Are you ready for an exciting challenge that will not only test your technical skills but also your creativity and problem-solving abilities?
Picture a scenario where communication is critical, but resources are limited. How would you tackle the task of positioning individuals efficiently, affordably, and in industrial environments with litle to no network coverage using Bluetooth/beacons on their wearables? This is the dilemma we want you to tackle in this thrilling competition.
Your goal is to develop an innovative solution that enables precise and efficient localization of individuals in a complex industrial environment, such as a large factory with few employees, where instant communication is essential for safety and operational efficiency. Imagine a machine that requires immediate attention, but there are only a few workers available. How can you ensure that the nearest worker is alerted quickly and accurately?
This challenge not only requires a deep understanding of technologies like the Internet of Things (IoT), Artificial Intelligence (AI), and cybersecurity but also a visionary outlook and the ability to think outside the box. We’re looking for teams that can create solutions that are not only technically robust but also practical and economically viable.
With your talent and creativity, you can be part of the technological revolution that will transform how we communicate and interact in industrial environments and beyond. Join us in this exciting challenge and leave your mark on the future of communication. Get ready to surpass your own expectations and inspire the world with your innovation.