Integration of enchanced jump point search (JPS) algorithm with modified bresenham technique for path planning in virtual grid-based environment
Jump Point Search is one of the path finding algorithm with huge advantage of maintaining zero memory overhead as no preprocessing process involved. However, despite of JPS advantage in using less memory, it does not consider the shape of obstacles when finding the shortest path from start to goal p...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/42802/1/24%20PAGES.pdf https://eprints.ums.edu.my/id/eprint/42802/2/FULLTEXT.pdf https://eprints.ums.edu.my/id/eprint/42802/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Jump Point Search is one of the path finding algorithm with huge advantage of maintaining zero memory overhead as no preprocessing process involved. However, despite of JPS advantage in using less memory, it does not consider the shape of obstacles when finding the shortest path from start to goal point. We demonstrate our framework; the integration of enhanced Jump Point Search algorithm with modified Bresenham for heuristic computation in virtual grid-based environment. The first approach is the enhancement of the JPS technique. Basically, the original JPS technique consists of two rules; pruning rules and jumping rules. This research study will be focusing on enhancement of jumping rules. The enhancement being done by improving the selection of jump point into n’+1, instead of choosing the node n’ next to the obstacle. The second approach is the modification of heuristic computation using original A* and modified Bresenham. Bresenham Line Algorithm is a line generation algorithm using integer arithmetic where the points (x1, y1) and (x2, y2) are assumed not equal and integer valued. The algorithm which consider the error value, e to increment pixels is the crucial part of the line generation applies the same concept in the second approach, where the error value, e is used as indicator to determine the heuristic cost of the path. The third approach is the integration of the two crucial process earlier which satisfy the third objective of the research study; to produce an optimal shortest path with minimum computation time. The visualization and experiment results from the enhanced JPS showed that the proposed technique require less number of nodes attached to obstacles as the obstacles level become higher in percentage. This is due to the improved jumping rules and steps of successor selection. The conducted simulations clearly shows that the proposed technique helps in producing a path with less nodes attached to the obstacles compared to the original JPS. Finally, the third approach of the integration of enhanced JPS with modified Bresenham technique is tested on three different benchmark maps; “Starcraft” map, Britannia height map and general height map. The results shows that this proposed technique performs better than the Original JPS manage to achieve shorter execution time while trying to maintain the shortest path towards reaching the goal. For example, in the last experiment part of using general height map, the execution time is 10ms with shortest path of 91 node distance compared to Original JPS with 40ms and 122 node distance. In conclusion, the integration of the two techniques has successfully produce an optimal shortest path with minimum execution time especially for environment with many obstacles. Next, future research can be further by enhancing the heuristic computation in JPS or other path finding technique by considering weightage heuristic cost. |
|---|