Adversarial Skill Learning for Robust Manipulation

Deep reinforcement learning has made significant progress in robotic manipulation tasks and it works well in the ideal disturbance-free environment. However, in a real-world environment, both internal and external disturbances are inevitable, thus the performance of the trained policy will dramatically drop. To improve the robustness of the policy, we introduce the adversarial training mechanism to the robotic manipulation tasks in this paper, and an adversarial skill learning algorithm based on soft actor-critic (SAC) is proposed for robust manipulation. Extensive experiments are conducted to demonstrate that the learned policy is robust to internal and external disturbances. Additionally, the proposed algorithm is evaluated in both the simulation environment and on the real robotic platform.

Authors

• Pingcheng Jian
• Chao Yang 0026
• Di Guo 0002
• Huaping Liu 0001
• Fuchun Sun 0001

Keywords

• Training
• Automation
• Conferences
• Reinforcement learning
• Robustness
• Task analysis
• Robots
• Generative Adversarial Networks
• Simulation Environment
• External Disturbances
• Deep Reinforcement Learning
• Real-world Environments
• Environmental Disturbances
• Adversarial Training
• Robotic Tasks
• Real Robot
• Training Policy
• Internal Disturbances
• Task In This Paper
• Protagonist
• Simulation Experiments
• Random Noise
• High Robustness
• Joint Space
• Real-world Experiments
• Robot Control
• Markov Decision Process
• Adversarial Attacks
• Reinforcement Learning Methods
• Critic Network
• Stochastic Policy
• Internal Noise
• Goal Position
• Robot Joint
• Fast Gradient Sign Method
• Random Action
• Types Of Disturbances