VMC-RRT: Virtual Motion Camouflage-Inspired Sampling-based Path Planning

In sample-based path planning algorithms, performing collision checking often requires a significant amount of time. This paper proposes a sample-based path planning algorithm called VMC-RRT, which is based on the Virtual Motion Camouflage phenomenon. Within the VMC framework, the algorithm reduces the dimensionality of obstacle information and introduces a new method for collision checking, which accelerates the speed of collision checking in sample-based algorithms. Additionally, the VMC framework provides better options for the generation of new nodes, allowing the VMC-RRT method to find a feasible solution from the start to the goal in fewer iterations. Moreover, the VMC-RRT approach can be combined with asymptotically optimal algorithms to gradually converge the feasible solution to the optimal one after it is found. Experiments show that the algorithm effectively reduces the time spent on collision checking and the total number of iterations, thus speeding up the process of finding a feasible solution in sample-based path planning algorithms. When combined with RRT*, the scheme can achieve an asymptotically optimal path, verifying the advantages of the proposed approach.