LES: Locally Exploitative Sampling for Robot Path Planning
This work addresses path planning for robots by offering an incremental improvement in sampling efficiency for specific domains.
The paper tackles the problem of robot path planning by introducing a locally exploitative sampling strategy that biases sampling towards improving local cost-to-come values, resulting in faster convergence to optimal solutions compared to state-of-the-art methods, as demonstrated in higher-dimensional robotic tasks.
Sampling-based algorithms solve the path planning problem by generating random samples in the search-space and incrementally growing a connectivity graph or a tree. Conventionally, the sampling strategy used in these algorithms is biased towards exploration to acquire information about the search-space. In contrast, this work proposes an optimization-based procedure that generates new samples to improve the cost-to-come value of vertices in a neighborhood. The application of proposed algorithm adds an exploitative-bias to sampling and results in a faster convergence to the optimal solution compared to other state-of-the-art sampling techniques. This is demonstrated using benchmarking experiments performed fora variety of higher dimensional robotic planning tasks.