Max Robotics — Global Service Robot Directory + US Certification

About

Ching-Yao Chan is a prominent researcher specializing in autonomous driving, deep reinforcement learning, and pedestrian trajectory prediction — core pillars of intelligent transportation systems. Based at the University of California, Berkeley, Chan has made significant contributions to solving some of the most complex challenges in vehicle automation and urban mobility. Chan's most influential work focuses on automated lane-changing behavior, where his team pioneered the application of Proximal Policy Optimization (PPO)-based deep reinforcement learning to enable vehicles to make safe, adaptive lane-change decisions without relying on rigid rule-based frameworks. This research has garnered over 120 citations, reflecting its substantial influence on the autonomous driving community. A companion study from 2019, exploring deep reinforcement learning for driving decision-making, has accumulated an additional 75 citations, cementing Chan's role as an early advocate for learning-based vehicular control. Beyond vehicle behavior, Chan has advanced pedestrian trajectory prediction through innovative graph-based models and the novel Pseudo-Oracle framework (TPPO), which addresses uncertainties in human motion and social interactions — work that has attracted nearly 70 citations. Collectively, Chan's research portfolio demonstrates a sustained commitment to making autonomous systems safer, smarter, and more socially aware, earning him recognition as a key voice in next-generation transportation research.