{"ID":5935840,"CreatedAt":"2026-07-07T01:22:02.77346169Z","UpdatedAt":"2026-07-07T02:10:06.972658124Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.03125","arxiv_id":"2607.03125","title":"Integrating Physics-Informed Neural Networks for Safe Reinforcement Learning in a 1-DoF Helicopter System","abstract":"Deep reinforcement learning (DRL) offers powerful control for industrial cyber-physical systems (ICPSs), but its \"black-box\" exploration risks violating strict hardware safety limits. Typically, these constraints are managed through complex reward shaping. In this work-in-progress paper, we embed a differentiable physics model directly into the proximal policy optimization (PPO) actor loss function. By simulating short-horizon future trajectories during training, the policy is penalized for anticipated safety violations independent of the task-reward signal. Evaluated on a simulated 1-degree-of-freedom helicopter testbed with strict pitch constraints, our physics-informed soft regularizations substantially reduce constraint violations while maintaining reliable target tracking.","short_abstract":"Deep reinforcement learning (DRL) offers powerful control for industrial cyber-physical systems (ICPSs), but its \"black-box\" exploration risks violating strict hardware safety limits. Typically, these constraints are managed through complex reward shaping. In this work-in-progress paper, we embed a differentiable physi...","url_abs":"https://arxiv.org/abs/2607.03125","url_pdf":"https://arxiv.org/pdf/2607.03125v1","authors":"[\"Georg Schäfer\",\"Jakob Rehrl\",\"Stefan Huber\"]","published":"2026-07-03T09:14:04Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.RO\"]","methods":"[\"Reinforcement Learning\",\"LoRA\"]","has_code":false}