Towards Physically Safe Reinforcement Learning under Supervision

2019-01-19 · Yinan Zhang, Devin Balkcom, Haoxiang Li

This paper addresses the question of how a previously available control policy $\pi_s$ can be used as a supervisor to more quickly and safely train a new learned control policy $\pi_L$ for a robot. A weighted average of the supervisor and learned policies is used during trials, with a heavier weight initially on the supervisor, in order to allow safe and useful physical trials while the learned policy is still ineffective. During the process, the weight is adjusted to favor the learned policy. As weights are adjusted, the learned network must compensate so as to give safe and reasonable outputs under the different weights. A pioneer network is introduced that pre-learns a policy that performs similarly to the current learned policy under the planned next step for new weights; this pioneer network then replaces the currently learned network in the next set of trials. Experiments in OpenAI Gym demonstrate the effectiveness of the proposed method.

📄 PDF Abstract BibTeX arXiv:1901.06576

Code (0)

등록된 구현이 없습니다.

Tasks

OpenAI Gym reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)Safe Reinforcement Learning

Weakly Supervised Reinforcement Learning for Autonomous Highway Driving via Virtual Safety Cages

2021-03-17 · Sampo Kuutti, Richard Bowden, Saber Fallah

The use of neural networks and reinforcement learning has become increasingly popular in autonomous vehicle control. However, the opaqueness of the resulting control policies presents a significant barrier to deploying n…

Autonomous Vehicles reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Beyond Self-Play: Hierarchical Reasoning for Continuous Motion in Closed-Loop Traffic Simulation

2026-05-09 · Weifan Zhang, Xiaofeng Zhao, Adel Bazzi, Mingrui Li 외 arxiv

Closed-loop traffic simulation requires agents that are both scalable and behaviorally realistic. Recent self-play reinforcement learning approaches demonstrate strong scalability, but their equilibrium strategies fail t…

Multi-agent Reinforcement Learning

Dynamic-TD3: A Novel Algorithm for UAV Path Planning with Dynamic Obstacle Trajectory Prediction

2026-04-30 · Wentao Chen, Jingtang Chen, Mingjian Fu, Tiantian Li 외 arxiv

Deep reinforcement learning (DRL) finds extensive application in autonomous drone navigation within complex, high-risk environments. However, its practical deployment faces a safety-exploration dilemma: soft penalty mech…

Reinforcement Learning Trajectory Prediction Collision Avoidance Drone navigation

Kinematics-Aware Latent World Models for Data-Efficient Autonomous Driving

2026-03-07 · Jiazhuo Li, Linjiang Cao, Qi Liu, Xi Xiong arxiv

Data-efficient learning remains a central challenge in autonomous driving due to the high cost and safety risks of large-scale real-world interaction. Although world-model-based reinforcement learning enables policy opti…

Reinforcement Learning Autonomous Driving

Physically-Grounded Goal Imagination: Physics-Informed Variational Autoencoder for Self-Supervised Reinforcement Learning

2025-11-10 · Lan Thi Ha Nguyen, Kien Ton Manh, Anh Do Duc, Nam Pham Hai arxiv

Self-supervised goal-conditioned reinforcement learning enables robots to autonomously acquire diverse skills without human supervision. However, a central challenge is the goal setting problem: robots must propose feasi…

Reinforcement Learning