| Popis: |
Rover missions exploring other planets are tightly constrained regarding the trade-off between safety and traversal speed. Detecting and avoiding hazards during navigation is capital to preserve the mobility of a rover. Low traversal speeds are often enforced to assure that wheeled rovers do not become stuck in challenging terrain, hindering the performance and scientific return of the mission. Even such precautions do not guarantee safe navigation due to non-geometric hazards hidden in the terrain, such as sand traps beneath thin duricrusts. These issues motivate the research of the interaction with rough and sandy planetary terrains of conventional and innovative robot locomotion concepts. Hybrid wheel-legs combine the mechanical and control simplicity of wheeled locomotion with the enhanced mobility of legged locomotion. This concept has been rarely proposed for planetary exploration and the study of its interaction with granular terrains is at a very early stage. This research focuses on advancing the state-of-the-art of wheel-leg-soil interaction analysis and applying it through in-situ sensing to simultaneously improve the speed and safety of planetary rover missions. The semi-empirical approach used combines both theoretical modelling and experimental analysis of data obtained in laboratory and field analogues. A novel light-weight, low-power sensor system, capable of reliably detecting wheel-leg sinkage and slippage phenomena on-the-fly, is designed, implemented and tested both as part of a simplified single-wheel-leg test bed and integrated in a fully mobile micro-rover. Moreover, existing analytical models for the interaction between deformable terrain and heavily-loaded wheels or lightly-loaded legs are adapted to the generalised medium-loaded multi-legged wheel-leg case and combined into hybrid approaches for better accuracy, as validated against experimental data. Finally, the soil sensor system and analytical models proposed are used to develop and prove the effectiveness of different solutions for soil characterisation, trafficability assessment and terrain classification based on non-geometric physical properties. |
