Dynamic Conditional Imitation Learning for Autonomous Driving
This work addresses critical safety and reliability issues in autonomous driving for real-world deployment, though it is incremental as it builds upon existing CIL methods.
The paper tackled the problem of conditional imitation learning (CIL) for autonomous driving, which suffers from poor generalization to unseen environments, inconsistency in varying weather, and inability to avoid static road blockages, by proposing a method that fuses laser scanner and camera data, introduces an efficient occupancy grid mapping, and adds algorithms for blockage avoidance and global route planning, resulting in improvements such as a 52% increase in success rate generalization and a 1.5 times increase in kilometers traveled before collision.
Conditional imitation learning (CIL) trains deep neural networks, in an end-to-end manner, to mimic human driving. This approach has demonstrated suitable vehicle control when following roads, avoiding obstacles, or taking specific turns at intersections to reach a destination. Unfortunately, performance dramatically decreases when deployed to unseen environments and is inconsistent against varying weather conditions. Most importantly, the current CIL fails to avoid static road blockages. In this work, we propose a solution to those deficiencies. First, we fuse the laser scanner with the regular camera streams, at the features level, to overcome the generalization and consistency challenges. Second, we introduce a new efficient Occupancy Grid Mapping (OGM) method along with new algorithms for road blockages avoidance and global route planning. Consequently, our proposed method dynamically detects partial and full road blockages, and guides the controlled vehicle to another route to reach the destination. Following the original CIL work, we demonstrated the effectiveness of our proposal on CARLA simulator urban driving benchmark. Our experiments showed that our model improved consistency against weather conditions by four times and autonomous driving success rate generalization by 52%. Furthermore, our global route planner improved the driving success rate by 37%. Our proposed road blockages avoidance algorithm improved the driving success rate by 27%. Finally, the average kilometers traveled before a collision with a static object increased by 1.5 times. The main source code can be reached at https://heshameraqi.github.io/dynamic_cil_autonomous_driving.