TR2024-004

Lunar Landing with Feasible Divert using Controllable Sets


    •  Srinivas, N., Vinod, A.P., Di Cairano, S., Weiss, A., "Lunar Landing with Feasible Divert using Controllable Sets", AIAA SciTech, January 2024.
      BibTeX TR2024-004 PDF
      • @inproceedings{Srinivas2024jan,
      • author = {Srinivas, Neeraj and Vinod, Abraham P. and Di Cairano, Stefano and Weiss, Avishai},
      • title = {Lunar Landing with Feasible Divert using Controllable Sets},
      • booktitle = {AIAA SciTech},
      • year = 2024,
      • month = jan,
      • url = {https://www.merl.com/publications/TR2024-004}
      • }
  • MERL Contacts:
  • Research Areas:

    Computer Vision, Control, Dynamical Systems, Optimization

Abstract:

We develop a guidance policy for a lunar lander under state and input constraints to land at a nominal target location, while maintaining the ability to divert to alternative landing sites if a hazard is detected. To ensure divert feasibility, we compute controllable sets that characterize the set of states that can be driven to a landing site for a range of fuel mass and length of divert trajectories. In the event a hazard is detected, we select the best divert feasible set among the landing targets according to a scoring metric of attributes such as fuel utilization and hazards. Simulations of the policy demonstrate that the lander can safely divert to an alternative landing site after a hazard is detected. We develop a visualization environment using Unreal Engine to render a landing trajectory at the lunar south pole using high fidelity digital elevation maps obtained by the Lunar Reconnaissance Orbiter. The visualization environment enables us to generate synthetic imagery of the lunar surface for terrain relative navigation and hazard detection.