Deep Learning Traversability Estimator for Mobile Robots in Unstructured Environments
This work addresses safe navigation for mobile robots in challenging terrains, but it is incremental as it builds on existing deep learning methods with application to specific robotic scenarios.
The paper tackles the problem of real-time terrain traversability analysis for mobile robots in unstructured environments by proposing a deep learning framework that estimates failure events from elevation maps and trajectories, achieving over 94% recall at 30% of the computational time compared to a simulator and demonstrating effective transfer to real-world Martian rover data.
Terrain traversability analysis plays a major role in ensuring safe robotic navigation in unstructured environments. However, real-time constraints frequently limit the accuracy of online tests especially in scenarios where realistic robot-terrain interactions are complex to model. In this context, we propose a deep learning framework trained in an end-to-end fashion from elevation maps and trajectories to estimate the occurrence of failure events. The network is first trained and tested in simulation over synthetic maps generated by the OpenSimplex algorithm. The prediction performance of the Deep Learning framework is illustrated by being able to retain over 94% recall of the original simulator at 30% of the computational time. Finally, the network is transferred and tested on real elevation maps collected by the SEEKER consortium during the Martian rover test trial in the Atacama desert in Chile. We show that transferring and fine-tuning of an application-independent pre-trained model retains better performance than training uniquely on scarcely available real data.