Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2606.00059 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918532385603584 |
|---|---|
| author | Langschwert, Julian Schaefer, Georg Rehrl, Jakob Huber, Stefan Hirlaender, Simon |
| author_facet | Langschwert, Julian Schaefer, Georg Rehrl, Jakob Huber, Stefan Hirlaender, Simon |
| contents | Informative excitation signals are critical for accurate system identification of mechatronic systems, yet classical system identification (SI) approaches require expert knowledge and hand-crafted signal design to respect hardware safety constraints, limiting their generalizability. We propose a reinforcement learning (RL) agent that learns optimal excitation signals for a Quanser Aero 2 testbed while autonomously enforcing safety constraints through reward shaping. Evaluated across 10 independent training seeds, our comprehensive agent achieves competitive estimation accuracy across all three identified parameters, outperforming classical baselines while incurring only 0.75% safety violations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_00059 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Reinforcement Learning for Optimal Experiment Design in Parameter Identification of Mechatronic Systems Langschwert, Julian Schaefer, Georg Rehrl, Jakob Huber, Stefan Hirlaender, Simon Robotics Machine Learning Informative excitation signals are critical for accurate system identification of mechatronic systems, yet classical system identification (SI) approaches require expert knowledge and hand-crafted signal design to respect hardware safety constraints, limiting their generalizability. We propose a reinforcement learning (RL) agent that learns optimal excitation signals for a Quanser Aero 2 testbed while autonomously enforcing safety constraints through reward shaping. Evaluated across 10 independent training seeds, our comprehensive agent achieves competitive estimation accuracy across all three identified parameters, outperforming classical baselines while incurring only 0.75% safety violations. |
| title | Reinforcement Learning for Optimal Experiment Design in Parameter Identification of Mechatronic Systems |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2606.00059 |