Dimension-variable Mapless Navigation with Deep Reinforcement Learning
This work addresses a specific limitation in robot navigation for robotics applications, enabling broader applicability of DRL methods, but it is incremental as it builds on existing DRL techniques.
The paper tackles the problem of deep reinforcement learning (DRL) agents being limited to fixed robot dimensions for mapless navigation, proposing a dimension-variable skill transfer method that allows robots with different dimensions to navigate successfully without retraining, as demonstrated in simulated and real-world experiments.
Deep reinforcement learning (DRL) has exhibited considerable promise in the training of control agents for mapless robot navigation. However, DRL-trained agents are limited to the specific robot dimensions used during training, hindering their applicability when the robot's dimension changes for task-specific requirements. To overcome this limitation, we propose a dimension-variable robot navigation method based on DRL. Our approach involves training a meta agent in simulation and subsequently transferring the meta skill to a dimension-varied robot using a technique called dimension-variable skill transfer (DVST). During the training phase, the meta agent for the meta robot learns self-navigation skills with DRL. In the skill-transfer phase, observations from the dimension-varied robot are scaled and transferred to the meta agent, and the resulting control policy is scaled back to the dimension-varied robot. Through extensive simulated and real-world experiments, we demonstrated that the dimension-varied robots could successfully navigate in unknown and dynamic environments without any retraining. The results show that our work substantially expands the applicability of DRL-based navigation methods, enabling them to be used on robots with different dimensions without the limitation of a fixed dimension. The video of our experiments can be found in the supplementary file.