Abstract
AbstractWe investigate how well a physics-based simulator can replicate a real wheel loader performing bucket filling in a pile of soil. The comparison is made using field-test time series of the vehicle motion and actuation forces, loaded mass, and total work. The vehicle was modeled as a rigid multibody system with frictional contacts, driveline, and linear actuators. For the soil, we tested discrete-element models of different resolutions, with and without multiscale acceleration. The spatiotemporal resolution ranged between 50–400 mm and 2–500 ms, and the computational speed was between 1/10,000 to 5 times faster than real time. The simulation-to-reality gap was found to be around 10% and exhibited a weak dependence on the level of fidelity, e.g., compatible with real-time simulation. Furthermore, the sensitivity of an optimized force-feedback controller under transfer between different simulation domains was investigated. The domain bias was observed to cause a performance reduction of 5% despite the domain gap being about 15%.
Publisher
Springer Science and Business Media LLC
Reference49 articles.
1. Allevato, A.D., Schaertl Short, E., Pryor, M., Thomaz, A.L.: Iterative residual tuning for system identification and sim-to-real robot learning. Auton. Robots 44, 1167–1182 (2020)
2. Aoshima, K., Fälldin, A., Wadbro, E., Servin, M.: World modeling for autonomous wheel loaders (2023). ArXiv preprint arXiv:2309.12016
3. Aoshima, K., Servin, M., Wadbro, E.: Simulation-based optimization of high-performance wheel loading. In: Feng, C., et al. (eds.) Proceedings of the 38th International Symposium on Automation and Robotics in Construction (ISARC), Dubai, UAE, November 2021, pp. 688–695. International Association for Automation and Robotics in Construction (IAARC) (2021)
4. Atkeson, C.G., Schaal, S.: Robot learning from demonstration. In: ICML, vol. 97, pp. 12–20 (1997)
5. Azulay, O., Shapiro, A.: Wheel loader scooping controller using deep reinforcement learning. IEEE Access 9, 24145–24154 (2021)