Abstract
A new heuristic searching algorithm based on fuzzy evaluation is proposed to solve the kick problem in RoboCup. With the help of discrete position of kickable area, fuzzy inference system is used to evaluate ball movement costs between two discrete positions. Accordingly, fuzzy rules are finally taken as knowledge to find a possible route, and searching algorithm is used to find the best kick sequence to reach the expected target velocity.
| Original language | English |
|---|---|
| Pages (from-to) | 571-575 |
| Number of pages | 5 |
| Journal | Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology |
| Volume | 23 |
| Issue number | 5 |
| Publication status | Published - Oct 2003 |
Keywords
- Fuzzy inference system
- Heuristic searching algorithm
- Reinforcement learning