Robot motion planning using adaptive hybrid sampling in probabilistic roadmaps

Abstract

Motion planning deals with finding a collision-free trajectory for a robot from the current position to the desired goal. For a high-dimensional space, sampling-based algorithms are widely used. Different sampling algorithms are used in different environments depending on the nature of the scenario and requirements of the problem. Here, we deal with the problems involving narrow corridors, i.e., in order to reach its destination the robot needs to pass through a region with a small free space. Common samplers used in the Probabilistic Roadmap are the uniform-based sampler, the obstacle-based sampler, maximum clearance-based sampler, and the Gaussian-based sampler. The individual samplers have their own advantages and disadvantages; therefore, in this paper, we propose to create a hybrid sampler that uses a combination of sampling techniques for motion planning. First, the contribution of each sampling technique is deterministically varied to create time efficient roadmaps. However, this approach has a limitation: The sampling strategy cannot adapt as per the changing configuration spaces. To overcome this limitation, the sampling strategy is extended by making the contribution of each sampler adaptive, i.e., the sampling ratios are determined on the basis of the nature of the environment. In this paper, we show that the resultant sampling strategy is better than commonly used sampling strategies in the Probabilistic Roadmap approach.