Vision-based Autonomous Driving for Unstructured Environments Using Imitation Learning
This addresses the problem of safe and smooth navigation in complex, lane-less environments for autonomous vehicles, representing an incremental improvement by applying imitation learning to a known bottleneck of noisy state recognition.
The paper tackles autonomous driving in unstructured environments by proposing a vision-based imitation learning method that directly drives toward drivable space using occupancy grid maps, avoiding the need for path planning and tracking. Experiments in real parking lots showed the method's effectiveness over model-based approaches, though specific numerical results were not provided.
Unstructured environments are difficult for autonomous driving. This is because various unknown obstacles are lied in drivable space without lanes, and its width and curvature change widely. In such complex environments, searching for a path in real-time is difficult. Also, inaccurate localization data reduce the path tracking accuracy, increasing the risk of collision. Instead of searching and tracking the path, an alternative approach has been proposed that reactively avoids obstacles in real-time. Some methods are available for tracking global path while avoiding obstacles using the candidate paths and the artificial potential field. However, these methods require heuristics to find specific parameters for handling various complex environments. In addition, it is difficult to track the global path accurately in practice because of inaccurate localization data. If the drivable space is not accurately recognized (i.e., noisy state), the vehicle may not smoothly drive or may collide with obstacles. In this study, a method in which the vehicle drives toward drivable space only using a vision-based occupancy grid map is proposed. The proposed method uses imitation learning, where a deep neural network is trained with expert driving data. The network can learn driving patterns suited for various complex and noisy situations because these situations are contained in the training data. Experiments with a vehicle in actual parking lots demonstrated the limitations of general model-based methods and the effectiveness of the proposed imitation learning method.