Author:
Muñoz Pablo,Bellutta Paolo,R-Moreno Maria D.
Abstract
AbstractThe on-ground operation of Mars rovers is a complex task that requires comprehensive planning in which path planning plays a fundamental role. The selection of paths has to be carefully chosen considering the scientific objectives, terrain, energy, and safety. In this regard, operators are assisted by path-planning algorithms that generate candidate paths based on cost functions. Distance traveled has always been considered one of the primary criteria when comparing paths. Other metrics such as the run-time to generate the solution or the number of expanded nodes are common measures considered in the literature. However, we want to analyze if those metrics provide useful information in challenging and partially known terrain. In this paper, we will review those metrics using two-path planning algorithms on real Mars maps. Based on our experience operating Mars rovers, we propose new metrics for assessing paths in real-world applications.
Funder
Junta de Comunidades de Castilla-La Mancha
Publisher
Springer Science and Business Media LLC
Reference29 articles.
1. Verma, V. et al. Autonomous robotics is driving Perseverance rover’s progress on Mars. Sci. Robot. 8, 3099 (2023).
2. Rothrock, B. et al. SPOC: Deep learning-based terrain classification for mars rover missions. AIAA Spacehttps://doi.org/10.2514/6.2016-5539 (2016).
3. Goh, E., Chen, J. & Wilson, B. Mars terrain segmentation with less labels. in 2022 IEEE Aerospace Conference (AERO), 1–10 (IEEE, 2022).
4. Raja, P. & Pugazhenthi, S. Optimal path planning of mobile robots: A reviews. Int. J. Phys. Sci. 7, 175–192. https://doi.org/10.5897/IJPS11.1745 (2012).
5. Ferguson, D. & Stentz, A. Field D*: An interpolation-based path planner and replanner. in Proc. of the International Symposium on Robotics Research (ISRR) (2005).