Collaborative Research: EAGER: CSR: Full-stack defense of vision-based autonomous driving systems

This project focuses on enhancing the safety and reliability of autonomous driving systems, making them more trustworthy in a variety of environments, including potentially hostile ones. The research will develop advanced detection and countermeasure techniques at the application, system, and hardware layers to help ensure that self-driving cars can make safe driving decisions even when faced with deliberate attempts to disrupt their operation. This involves creating robust deep-learning modules for driving decisions, enhancing the safety and security of automotive processors, and developing tools to detect and fix hardware issues in mission mode. By conducting extensive real-world testing with popular automotive benchmarks, the project aims to validate these innovations, ensuring they can be confidently adopted in everyday use. This comprehensive approach addresses current and future challenges in autonomous driving technology, paving the way for safer and more efficient transportation systems that the public can trust.

The broader impact of this research extends beyond automotive technology to other critical systems such as space exploration, smart medical devices, and various Internet of Things (IoT) applications. With the deep learning market expected to grow to $24.5 billion by 2025, advancements from this project will play a crucial role in ensuring the safety and security of numerous technologies that impact daily life. The project’s findings will be widely disseminated through publications, software releases, and educational courses, contributing to the development of a skilled workforce in this rapidly evolving field. Further, the researchers will engage in educational outreach, including workshops and summer camps for students from diverse backgrounds, promoting inclusivity and inspiring future scientists and engineers. The researchers’ ongoing collaboration with leading semiconductor companies will also support translation into practice. The project not only advances technology, but contributes to societal well-being by making autonomous systems safer and more reliable, ensuring that the benefits of these innovations are widely accessible.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Award Number: 2537757
Principal Investigator: Kanad Basu
Funds Obligated: $91,988
State: NY