Optimal Probabilistic Motion Planning with Partially Infeasible LTL Constraints.
Mingyu Cai, Zhijun Li, Hongbo Gao, Shaoping Xiao, Zhen Kan · arXiv (Cornell University) · 2020
This paper studies optimal probabilistic motion planning of a mobile agent with potentially infeasible task specifications subject to motion and environment uncertainties. Instead of the traditional Rabin automata, limit-deterministic Buchi automata are applied and a relaxed product MDP between PL-MDP (i.e., probabilistic labeled Markov decision process) and LDBA is developed, which allows the agent to revise its motion plan whenever the task is not fully feasible. A multi-objective optimization problem is then formulated to jointly consider the probability of the task satisfaction , the violation cost, and the implementation costs, which is solved via coupled linear programs. To the best of our knowledge, it is the first work that bridges the gap between planning revision and optimal control synthesis of both prefix and suffix of the agent trajectory. Simulation results are provided to demonstrate the effectiveness of the proposed framework.