Abstract | ||
---|---|---|
We present a learning algorithm designed to improve robot path planning. The algorithm relies on an existing path planner to provide solutions to difficult tasks. From these solutions, it learns a sparse network of useful robot subgoals that guides and supports fast planning. We analyze the algorithm theoretically by developing some general techniques useful in characterizing behaviors of probabilistic learning. We demonstrate the effectiveness of the algorithm empirically with an existing path planner in practical environments. The learning algorithm not only reduces the time cost of existing planners, but also increases their capability in solving difficult tasks. |
Year | DOI | Venue |
---|---|---|
1992 | 10.1016/B978-1-55860-247-2.50013-9 | ML |
Keywords | Field | DocType |
improving path planning,path planning | Motion planning,Any-angle path planning,Computer science,Robot path planning,Planner,Artificial intelligence,Probabilistic logic,Robot,Machine learning | Conference |
Issue | ISBN | Citations |
1 | 1-5586-247-X | 11 |
PageRank | References | Authors |
6.53 | 3 | 1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Pang C. Chen | 1 | 85 | 20.60 |