His current research is focused on the discovery of fundamental principles and the analysis and design of protocols/systems for next-generation vehicular networks, for safety, telematics and infotainment applications. degrees in electrical engineering, from University of Southern California, Los Angeles, in 2005. degree in automation engineering from Tsinghua University, Beijing, China, in 1999, and the M.S.E.E. Before joining General Motors research lab, he received the B.S. Fan Bai is a staff researcher and lab group manager in the Electrical & Control Systems Lab., Research & Development and Planning, General Motors Corporation. We also illustrate the critical architecture changes on both vehicle side and cloud side.ĭr. Using two example real-world applications – Simultaneous Localization and Mapping (SLAM) and collaborative perception enhancement, we explore how these applications can instead leverage cloud servers and edge servers, utilizing their inexpensive and elastic resource pool to seamlessly augment vehicle onboard computing capability. We propose MABSTA, a novel algorithm that learns the performance of unknown devices and channel qualities continually through exploratory probing and makes task assignment decisions by exploiting the gained knowledge. Later, we further enhance and formulate the task assignment problem as an online learning problem using an adversarial multi-armed bandit framework. First, we formulate an NP-hard problem to minimize the application latency while meeting prescribed resource utilization constraints and propose a novel fully polynomial time approximation scheme (FPTAS) to solve it. It is of our interest to find optimal assignments of tasks to local and remote devices that can take into account the application-specific profile, availability of computational resources, and link connectivity, and find a balance between energy consumption costs of mobile devices and latency for delay-sensitive applications.
To process this data and present information to the driver in real-time, in this talk, we introduce a real-time vehicle/cloud coordination system and a couple of novel online scheduling algorithms designed to exploit distributed devices that are connected via wireless links. In the future, we imagine physical infrastructure capable of sensing and communicating data to vehicles to improve a driver’s awareness on the road.