Evaluation of Safety Constraints in Autonomous Navigation with Deep Reinforcement Learning
This work addresses safety constraints for autonomous vehicles, but it is incremental as it builds on existing reinforcement learning methods.
The study tackled the problem of ensuring safety in autonomous navigation by comparing safe and unsafe reinforcement learning policies, finding that the safe policy increased obstacle clearance by 15% and reduced collisions by 30% during training without compromising overall performance.
While reinforcement learning algorithms have had great success in the field of autonomous navigation, they cannot be straightforwardly applied to the real autonomous systems without considering the safety constraints. The later are crucial to avoid unsafe behaviors of the autonomous vehicle on the road. To highlight the importance of these constraints, in this study, we compare two learnable navigation policies: safe and unsafe. The safe policy takes the constraints into account, while the other does not. We show that the safe policy is able to generate trajectories with more clearance (distance to the obstacles) and makes less collisions while training without sacrificing the overall performance.