Authors
Daniel M Gaines, Gregg R Rabideau, Stephen R Kuhn, Steve A Chien, Elyse Moffi, Edward Benowitz, Vincent Wong
Published in
Science robotics. Volume 11. Issue 119. Pages eaeh2947. Oct 28, 2026. Epub Oct 28, 2026.
Abstract
Operators of planetary rovers, under strict energy and thermal constraints, face substantial challenges in maintaining high levels of rover productivity because of conservative ground-based planning and long communication latencies. In October 2023, to overcome these limits and increase science return, the Mars 2020 mission deployed OnBoard Planner into Perseverance rover operations. OnBoard Planner uses artificial intelligence constraint-based search to dynamically construct and modify activity schedules on board, adapting to real-time changes in rover state and martian environmental conditions. By replacing conservative ground predictions with real-time resource tracking, the system dynamically adjusts heating durations, reclaims unused schedule margin, and inserts opportunistic activities when extra resources become available. Here, we report on OnBoard Planner's performance on Mars across more than 2 years of surface operations. The results demonstrate that OnBoard Planner increased discretionary rover activity by 10.5% overall while also improving the thermal and temporal conditions for certain types of scientific observations. Furthermore, the system achieved these productivity gains without increasing ground operations timelines. These findings are part of why OnBoard Planner has successfully become a standard part of Mars 2020 operations and demonstrate how dynamic onboard autonomy can enable resource-efficient operations for complex robotic systems in deep-space environments.
PMID:
42950063
Bibliographic data and abstract were imported from PubMed on 29 Oct 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 72
- Comments 0