CRII: CSR: Towards an Edge-enabled Software-Defined Vehicle Framework for Dynamic Over-the-Air Updates

Software-defined vehicles (SDVs) represent the next evolution of automotive development. Traditional methods for over-the-air (OTA) software updates in consumer electronics are not well-suited to SDVs due to differences in hardware complexity, software intricacy, and other factors. This work addresses the overlooked need for dynamic OTA updates and data transmission in SDVs. It offers a suite of systems, algorithms, and communication strategies to enhance SDV frameworks from the perspective of software downlink and data uplink. Specifically, this work will develop a vehicle-edge-cloud framework for SDVs, aiming to foster resilient collaborations between SDVs and roadside units (RSUs) to deliver verified software in a timely fashion. It will focus on efficient SDV data uplink while in motion, employing compressive sensing technologies for low-bandwidth, low-latency data transmission, thereby enhancing the practical deployment of SDVs. This work will introduce new strategies based on physical testbeds for (1) an open-source SDV framework and (2) dynamic OTA updates. If successful, it will enable real-time OTA updates to prevent accidents, save lives, and reduce costs due to software malfunctions. It will also support emergency response, adaptive regulatory compliance across state borders, and geolocation-based performance optimization. Moreover, it will support efficient and adaptive data transmission on the go, keeping vehicle software up-to-date. It has the potential to create research and education opportunities focused on widely distributed vehicles, robots, and edge devices throughout their lifecycles. These are essential elements to embrace the connected, autonomous, shared, and electric (CASE) trends in the digital era. 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.