JPL Technical Report Server

Multi-Agent Motion Planning using Deep Learning for Space Applications

Show simple item record

dc.contributor.author Yun, Kyongsik
dc.contributor.author Choi, Changrak
dc.contributor.author Alimo, Ryan
dc.contributor.author Davis, Anthony
dc.contributor.author Forster, Linda
dc.contributor.author Rahmani, Amir
dc.contributor.author Adil, Muhammad
dc.contributor.author Madani, Ramtin
dc.date.accessioned 2022-04-12T00:02:29Z
dc.date.available 2022-04-12T00:02:29Z
dc.date.issued 2020-11-16
dc.identifier.citation AIAA-ASCEND, Las Vegas, Nevada, November 16-19, 2020
dc.identifier.clearanceno CL#21-1233
dc.identifier.uri http://hdl.handle.net/2014/54577
dc.description.abstract State-of-the-art motion planners cannot scale to a large number of systems. Motion planning for multiple agents is an NP (non-deterministic polynomial-time) hard problem, so the computation time increases exponentially with each addition of agents. This computational demand is a major stumbling block to the motion planner's application to future NASA missions involving the swarm of space vehicles. We applied a deep neural network to transform computationally demanding mathematical motion planning problems into deep learning-based numerical problems. We showed optimal motion trajectories can be accurately replicated using deep learning-based numerical models in several 2D and 3D systems with multiple agents. The deep learning-based numerical model demonstrates superior computational efficiency with plans generated 1000 times faster than the mathematical model counterpart.
dc.description.sponsorship NASA/JPL en_US
dc.language.iso en_US
dc.publisher Pasadena, CA: Jet Propulsion Laboratory, National Aeronautics and Space Administration, 2020
dc.title Multi-Agent Motion Planning using Deep Learning for Space Applications
dc.type Preprint


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search


Browse

My Account