Formulation of an alternative rapidly-exploring random trees (RRT) sampling based algorithm through parameter alterations
Rapidly-exploring Random Trees (RRT) is one of the coveted algorithms for path planning. However, the said algorithm, including its variants are yet to be evaluated in environments with complex topologies and constraints. Specific suggestions include changing parameters such as step size and radius,...
Saved in:
| Main Author: | |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Animo Repository
2021
|
| Subjects: | |
| Online Access: | https://animorepository.dlsu.edu.ph/etdm_ece/6 https://animorepository.dlsu.edu.ph/cgi/viewcontent.cgi?article=1005&context=etdm_ece |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | De La Salle University |
| Language: | English |
| Summary: | Rapidly-exploring Random Trees (RRT) is one of the coveted algorithms for path planning. However, the said algorithm, including its variants are yet to be evaluated in environments with complex topologies and constraints. Specific suggestions include changing parameters such as step size and radius, as well as switching to other local planners to see positive effects and possible improvements. In this study, another RRT variant is formulated, named as RRT-M. Considering all necessary prerequisites, hardware and software requirements, mapping and localization, costmap configurations, and setting up the algorithm as a global planner plugin, experimentations were conducted in three map ennvironments (maze, bookstore, small village). Results show that RRT-M is compared to the RRT* base algorithm: at most a 61.983% improvement in path length, 58.4414% improvement in navigation duration, and 27.1768% in planning time. Through the produced graphs and visualizations, a qualitative assessment concludes that RRT-M works properly in narrow paths and prioritizes shorter path alternatives. |
|---|