Danger-aware Adaptive Composition of DRL Agents for Self-navigation
This work addresses the problem of efficient and safe robot navigation for mobile robots, offering a novel composition approach that is incremental in leveraging existing DRL methods.
The paper tackles the challenge of training robots to simultaneously learn goal-reaching and obstacle-avoidance skills for self-navigation by proposing a danger-aware adaptive composition (DAAC) framework that combines two pre-trained DRL agents without retraining, enabling robots to safely and quickly navigate unknown, complex environments.
Self-navigation, referred as the capability of automatically reaching the goal while avoiding collisions with obstacles, is a fundamental skill required for mobile robots. Recently, deep reinforcement learning (DRL) has shown great potential in the development of robot navigation algorithms. However, it is still difficult to train the robot to learn goal-reaching and obstacle-avoidance skills simultaneously. On the other hand, although many DRL-based obstacle-avoidance algorithms are proposed, few of them are reused for more complex navigation tasks. In this paper, a novel danger-aware adaptive composition (DAAC) framework is proposed to combine two individually DRL-trained agents, obstacle-avoidance and goal-reaching, to construct a navigation agent without any redesigning and retraining. The key to this adaptive composition approach is that the value function outputted by the obstacle-avoidance agent serves as an indicator for evaluating the risk level of the current situation, which in turn determines the contribution of these two agents for the next move. Simulation and real-world testing results show that the composed Navigation network can control the robot to accomplish difficult navigation tasks, e.g., reaching a series of successive goals in an unknown and complex environment safely and quickly.