Complex robot behaviors are often structured as state machines, where states encapsulate actions and a transition function switches between states. Since transitions depend on physical parameters, when the environment changes, a roboticist has to painstakingly readjust the parameters to work in the new environment. We present interactive SMT- based Robot Transition Repair (SRTR): instead of manually adjusting parameters, we ask the roboti- cist to identify a few instances where the robot is in a wrong state and what the right state should be. A lightweight automated analysis of the transition function’s source code then 1) identifies adjustable parameters, 2) converts the transition function into a system of logical constraints, and 3) formulates the constraints and user-supplied corrections as a MaxSMT problem that yields new parameter values. Our evaluation shows that SRTR is effective on real robots and in simulation. We show that SRTR finds new parameters 1) quickly, 2) with only a few corrections, and 3) that the parameters generalize to new scenarios. We also show that a simple state machine corrected by SRTR can out- perform a more complex, expert-tuned state machine in the real world.
Interactive Robot Transition Repair with SMT
Jarrett Holtz, Arjun Guha, and Joydeep Biswas. In Proceedings of IJCAI-ECAI-2018, the International Joint Conference on Artificial Intelligence Supported in part by AFRL and DARPA agreement #FA8750-16-2-0042, and by NSF grant CCF-1717636