Can I read Robust Lane Change via Adversarial RL on EtoBox?

Robust Lane Change via Adversarial RL by deepak suresh is a document available to read on EtoBox.

What is Robust Lane Change via Adversarial RL about?

This paper proposes an observation adversarial reinforcement learning (OARL) approach to improve the robustness of lane change decision making for autonomous vehicles under perception uncertainties. A constrained observation-robust Markov decision process is used to model lane change behaviors considering policy constraints and observation perturbations. An adversarial attack technique approximates optimal observation perturbations efficiently. A constrained observation-robust actor-critic algorithm optimiz

Author

deepak suresh

Language

EN