Sayed M.H. Jafri and Rahul Kala
Robot motion planning, outdoor path planning, R∗δ Risky algorithm,ε-admissible A∗ε, graph search, Gaussian process Bayesian classifier
The paper presents a work on the path planning of a mobile robot in the outdoor environment by using the A∗ algorithm. The algorithm searches for a path with a short path length and high safety, while being computationally efficient. The A∗ algorithm is operated on a probabilistic costmap which incorporates the uncertainties in the vision and mapping algorithms. To trade off between computation time and path cost, we used the A∗ ε algorithm, which operates in higher resolutions (of actions considered in expansions). The algorithm is then extended to incorporate uncertainties in the costmap, which represents the information about the environment in the form of a probability distribution. To make the A∗ algorithm compatible with a probabilistic costmap, we used the R∗ δ algorithm. In the resultant algorithm, we not only achieve the safety of the path, but we also decrease the execution time and the cost of the path.
Important Links:
Go Back
