Additional Project Information
In the evolving landscape of urban transportation, ensuring the safety of Vulnerable Road Users (VRUs) - pedestrians, bicyclists, and construction workers - has become imperative. Current approaches like Advanced Driving Assistance Systems (ADAS) and Co-operative Collision Avoidance Systems (CCAS) offer partial solutions but have limitations, such as reliance on direct participation from VRUs or being constrained by visibility and weather conditions. This project aims to address these challenges by developing a comprehensive, low-cost VRU safety system, integrating passive detection, sensor fusion, and advanced communication and security technologies.
|
Collaborative and Interdisciplinary Approach |
Leveraging the interdisciplinary expertise of multiple research teams, this project proposes a collaborative framework to develop technologies that address the safety needs of different VRU communities. By sharing resources and insights across teams, the project aims to become a leading voice in smart infrastructure for VRU safety, maximizing impact and fostering innovative solutions. |
|
Research Components |
Pedestrian Early Detection and Localization System (PEDALS): PEDALS focuses on enhancing VRU safety in urban settings. By integrating various sensors (camera, radar, LiDAR) and detecting Signals of Opportunity (SoOP) from devices like smartphones, this system aims to localize VRUs and communicate critical information to nearby vehicles. This proactive approach will allow for early warning systems and reduce collision risks. |
| Construction Safety Monitoring System (CoSMoS): Addressing the high rate of construction worker fatalities, CoSMoS is a specialized system for transportation infrastructure construction sites. Utilizing advanced computer vision and a deep learning model (Faster-R-CNN), this system will identify potential hazards in real-time, ensuring timely alerts and increasing on-site safety. | |
| Security for Autonomous Transportation Systems: With the rise of autonomous vehicles, addressing cybersecurity threats becomes crucial. This component of the project will develop robust defense mechanisms against common cyberattacks like jamming and relay attacks on Controller Area Network (CAN bus) and sensor networks in vehicles. The deployment of a hybrid convolutional/recurrent neural network (CRNN) will enhance the security and safety of autonomous vehicles on the road. |
Conclusion: This project represents a comprehensive effort to increase the safety of VRUs in urban environments. By combining advanced sensing technologies, intelligent data processing, and robust security measures, it aims to significantly reduce the risk of accidents and fatalities involving VRUs. The collaborative and interdisciplinary nature of the project ensures a holistic approach, promising substantial advancements in the field of Intelligent Transport Systems (ITS).