Towards Safe and Resilient Autonomy in Multi-Robot Systems

Autonomous systems are envisioned to increasingly co-exist with humans in our daily lives, from household service to large-scale warehouse logistics, agriculture environment sampling, and smart city. Among them, networked cooperative systems such as autonomous multi-robot systems have been widely studied given their capability of accomplishing complex tasks through cooperative behaviors. Reliable interactions among robots as networked safety-critical systems often require provably correct guarantees about safety (e.g. collision avoidance) and resilience (e.g. capability of maintaining communication and operating in an unknown environment). As we strive to design and control such a large-scale system, robots are often assumed to have perfect information (e.g. ground-truth state, system dynamics, and environment model information), unconstrained inter-robot communication, and fault-free operation. However, the precomputed guarantees based on these assumptions could be easily broken when deploying robots in the real world that is uncertain, rapidly changing, and inherently stochastic. In this thesis, we seek to develop and validate mathematically grounded algorithms to assure safe and resilient interactions among robots that adapt to uncertain and possibly hostile dynamic environments. To achieve the design objective, we discuss three research topics, including (1) safe control and learning under uncertainty, (2) resilient multi-robot interaction through networking, and (3) data-driven multi-robot coordination adapting to the unknown environment. For (1), we first propose a reactive safe control framework for multi-robot sys-First and foremost, I would like to express my sincere gratitude to my advisor Katia Sycara, who has been incredibly supportive throughout my journey at Carnegie Mellon University. Her profound knowledge, insightful visions, and patience not only helped me survive through struggling phases of PhD study and career development, but also greatly influenced me on how to keep passionate and be a deep thinker. It was always amazing to see how great ideas came from simple intuition during our numerous inspiring discussions. I feel so lucky to have the freedom to explore my own research ideas and stay happy when pursuing my goals. I sincerely wish I could be a great researcher and mentor as she is in the future. I am also grateful to my thesis committee members: Maxim Likhachev, Changliu Liu, and Amanda Prorok, who have provided extremely helpful discussion and feedback. Their vision, guidance, and knowledge have been invaluable in my work and this thesis. In addition, I am thankful to Maxim Likhachev for his strongest support throughout my Master and PhD study in supervising my research progress and providing tremendous help with great advice, without which I would not have been where I am today. I am fortunate to have spent two wonderful summers with Ashish Kapoor as a research intern at Microsoft. I learned a lot from the fruitful discussions with him that shaped my research in robot safety as one of the main parts of this thesis work. His insight, encouragement, and unique viewpoint on research and life have always been an inspiration for me. I am also thankful to my collaborators Nilanjan Chakraborty and Wen Sun, for always being available for many fruitful discussions that lead me to new research paths. I am also thankful to other faculty members at CMU and Pitt, especially