The global path planning of planetary surface rovers is crucial for optimizing exploration benefits and system safety. For the cases of long-range roving or obstacle constraints that are time-varied, there is an urgent need to improve the computational efficiency of path planning. This paper proposes a learning-based global path planning method that outperforms conventional searching and sampling-based methods in terms of planning speed. First, a distinguishable feature map is constructed through a traversability analysis of the extraterrestrial digital elevation model. Then, considering planning efficiency and adaptability, a hierarchical framework consisting of step iteration and block iteration is designed. For the planning of each step, an end-to-end step planner named SP-ResNet is proposed that is based on deep reinforcement learning. This step planner employs a double-branch residual network for action value estimation, and is trained over a simulated DEM map collection. Comparative analyses with baselines demonstrate the prominent advantage of our method in terms of planning speed. Finally, the method is verified to be effective on real lunar terrains using CE2TMap2015.

No abstract

No abstract

The use of mobile robots for assisting astronauts in extravehicular activities could be an effective option for improving mission productivity and crew safety. It is thus critical that these robots follow the astronaut and maintain a stable distance to provide personalized and timely assistance. However, most extraterrestrial bodies exhibit rugged terrain that can impede a robot’s movements. As such, a novel predictive-guide following strategy is proposed to improve the stability of astronaut–robot distance in obstructive environments. This strategy combines a deep reinforcement learning navigator and a Kalman filter-based predictor to generate optimized motion sequences for safely following the astronaut and acquire predictive guidance concerning future astronaut movements. The proposed model achieved a success rate of 95.0% in simulated navigation tasks and adapted well to untrained complex environments and varied robot movement settings. Comparative tests indicated our strategy managed to stabilize the following distance to within ±1.0 m of the reference value in obstructed environments, significantly outperforming other following strategies. The feasibility and advantage of the proposed approach was validated with a physical robotic follower in a Mars-like environment. [Formula: see text]