Safe Robot Autonomy in Unknown Environments

Recent advances in machine learning, computer vision, and control theory offer a tremendous opportunity to deploy autonomous robot systems to uncharted environments to accomplish complex missions. Such tasks are particularly challenging as they require the robots to operate in environments with unknown structure, degraded environmental conditions, severe communication and sensing constraints, and expansive areas of operation. With the goal of ultimately designing safe, robust and autonomous robotic systems that can operate in these conditions, in this talk, I will present a novel perception-based control framework that allows autonomous multi-robot systems to accomplish complex missions in unknown environments. The proposed approach generates reactive control policies that adapt to the continuously learned map of the environment that is updated using sensing and perception systems while it scales well with the number of robots and the size of workspace. Finally, it has been well studied that learning-based perception systems lack robustness rendering perception-based control loops vulnerable to adversarial attacks (e.g., adversarial stickers on stop signs). If time permits, I will briefly present a new attack agnostic and computationally light defense mechanism against adversarial attacks to perception systems.