Differentiable Constrained Imitation Learning for Robot Motion Planning and Control
![[Translate to English:] [Translate to English:]](/storages/rst-etit/_processed_/e/6/csm_large_language_12b3096fb6.png)
In the age of large language models, imposing constraints on learning-based approaches becomes even more critical. Therefore, we developed a method to integrate safety constraints into imitation learning approaches for general robotic applications.