Reactive Behavioral Strategy for Unmanned Ground Vehicle Under Liner Temporal Logic Specifications
Liangguo Liu, Jun Peng, Xiaoyong Zhang, Rui Zhang, Bin Chen, Kai Gao, Yingze Yang · 2018
Making the feasible and safe behavioral strategy for unmanned ground vehicle in urban traffic environment is a challenging issue, due to various traffic rules, complicated road situations and changeable traffic conditions. In this paper, a method that allows reactive behavioral strategy for unmanned ground vehicle is proposed. The unmanned vehicle is modeled as a reactive system which keeps interacting with its environment. The reactive system has two subsystems of transition system and Nondeterministic Büchi Automaton. The motion and behaviors of the unmanned ground vehicle in the environment are regarded as a transition system. The sensor specifications that depend on the interacting information data between surroundings and the unmanned ground vehicle are treated as the trigger condition of the transition system. Traffic rules, topological structure constraints of road and empirical decision are described with the fragments of liner temporal logic formulas. The liner temporal logic formulas are converted to the Nondeterministic Büchi Automaton. The rationality of the results obtained from transition system is checked by Nondeterministic Büchi Automaton, which limits the behaviors of the unmanned vehicle satisfy liner temporal logic specifications. Thus, the method results in a expected behavioral strategy satisfying the high-level specification under the admissible environment. The experiment is designed and completed by using the Liner Temporal Logic MissiOn Planning toolkit and the results illustrate its performance.