Multi-Agents Spatial Visibility Trajectory Planning and Patrolling Using Inverse Reinforcement Learning

In this paper, we present a conceptual Spatial
Trajectory Planning (STP) method using Rapid Random Trees
(RRT) planner, generating visibility motion primitives in
urban environments using Inverse Reinforcement Learning
(IRL) approach. Visibility motion primitives are set by using
Spatial Visibility Clustering (SVC) analysis. Based on the STP
planning method, we introduce IRL formulation and analysis
which learns the value function of the planner from
demonstrated trajectories and generating spatial visibility
trajectory planning. Additionally, we study the visible
trajectories planning for patrolling application using
heterogeneous multi agents in 3D urban environments. Our
concept is based on spatial clustering method using visibility
analysis of the 3D visibility problem from a viewpoints in 3D
urban environments, defined as locations. We consider two
kinds of agents, with different kinematic and perception
capabilities. Using simplified version of Traveling Salesman
Problem (TSP), we formulate the problem as patrolling
strategy one, with upper bound optimal performances. We
present a combination of relative deadline UniPartition
approaches based on visibility clusters. These key features
allow new planning optimal patrolling strategy for
heterogeneous agents in urban environment. We demonstrate
our patrolling strategy method in simulations using
Autonomous Navigation and Virtual Environment Laboratory
(ANVEL) test bed environment.