DRL-Based Pose Control for Double-Ackermann Robots Under Actuation Uncertainties
For researchers deploying DRL on non-holonomic mobile robots, this work provides a practical sim-to-sim-to-real approach to mitigate actuation uncertainty, though it is an incremental extension of an existing framework.
The paper extends a DRL-based pose control framework for double-Ackermann robots and addresses sim-to-real transfer issues due to actuation uncertainties. By incorporating actuation effects from a high-fidelity simulator into training, they improved success rate from 25% to 92% in Gazebo and achieved 69% under stricter thresholds, with successful real-robot transfer.
Robust deployment of deep reinforcement learning (DRL) policies on real robots remains challenging due to discrepancies between simulation and real-world dynamics. We address this issue in the context of maneuvering with double-Ackermann-steering mobile robots, which introduce additional constraints due to their non-holonomic nature. Building upon the DRL framework ManeuverNet, we extend its objective from position control to full pose control, resulting in a more challenging task. We further investigate the impact of actuation-related uncertainties on policy transfer. The use of simplified actuation models during training of the extended policy can lead to poor generalization, shown by a success rate drop from 100% in PyBullet to 25% in Gazebo under stricter evaluation conditions. To address this limitation, we adopt a sim-to-sim-to-real approach, where actuation effects observed in Gazebo are incorporated into the PyBullet training environment. Using multi-environment DRL with SAC and CrossQ, we learn policies that remain robust despite modeling inaccuracies. This approach can significantly reduce the performance gap across simulators, achieving up to 92% success rate in Gazebo and maintaining 69% under stricter thresholds, with successful transfer to a real robot without additional tuning.