MAMPS: Safe Multi-Agent Reinforcement Learning via Model Predictive Shielding

October 25, 2019 Β· Declared Dead Β· πŸ› arXiv.org

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Wenbo Zhang, Osbert Bastani, Vijay Kumar arXiv ID 1910.12639 Category eess.SY: Systems & Control (EE) Cross-listed cs.AI, cs.MA, cs.RO Citations 43 Venue arXiv.org Last Checked 6 months ago
Abstract
Reinforcement learning is a promising approach to learning control policies for performing complex multi-agent robotics tasks. However, a policy learned in simulation often fails to guarantee even simple safety properties such as obstacle avoidance. To ensure safety, we propose multi-agent model predictive shielding (MAMPS), an algorithm that provably guarantees safety for an arbitrary learned policy. In particular, it operates by using the learned policy as often as possible, but instead uses a backup policy in cases where it cannot guarantee the safety of the learned policy. Using a multi-agent simulation environment, we show how MAMPS can achieve good performance while ensuring safety.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Systems & Control (EE)

Died the same way β€” πŸ‘» Ghosted