Coulomb force-guided deep reinforcement learning for effective and explainable robotic motion planning
Sirui Song, Trevor Bihl, Jundong Liu · 2026
Read Coulomb force-guided deep reinforcement learning for effective and explainable robotic motion planning by Sirui Song, Trevor Bihl, Jundong Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12900773.txt.
Description
Sirui Song and co-authors guide robot motion with a physics trick drawn from charge forces. The study appeared in 2025. The robot, its goal, and blocks in its path are cast as charges. Pull and push forces among them shape the reward the learner chases.
Edge lines of blocks, read from laser scans, add early rewards that steer the robot off blocks from afar. The model first learned in a sim world, then ran on a real small robot. Tests in both worlds showed far fewer hits than past methods. Paths kept safe gaps from blocks and reached goals on safer lines. The force frame also makes each choice plain to read.
Catalog record
| CREATOR | Sirui Song, Trevor Bihl, Jundong Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |