Show simple item record

dc.contributor.authorWalton, Claire
dc.contributor.otherCenter for Autonomous Vehicle Research (CAVR)
dc.date4/17/2018
dc.date.accessioned2018-05-04T00:08:37Z
dc.date.available2018-05-04T00:08:37Z
dc.date.issued2018-04-17
dc.identifier.urihttp://hdl.handle.net/10945/58058
dc.descriptionCRUSER TechCon 2018 Research at NPS
dc.description.abstractUnmanned vehicles have demonstrated their utility in every operational domain: land, sea, air, and space. Heterogeneous vehicles and sensors can be combined to form mobile sensing and communications networks that leverage their individual capabilities, but these networks must be responsive to rapidly changing events across multiple domains. Prior ONR- and CRUSER-funded research at NPS has produced new algorithms and computational tools for solving multi-agent optimization problems under time or resource constraints. These tools incorporate vehicle dynamics, sensor characteristics, and probabilistic models to plan vehicle trajectories which optimize mission objectives. These algorithms have been successfully applied to solve optimal search problems during mine countermeasures (MCM) or intelligence, surveillance, and reconnaissance (ISR) operations. This capability was recently demonstrated at CRUSER's Multi-Thread Experiment (MTX) on San Clemente Island in November 2018. At present, optimal mission plans are computed off-line, prior to launch, and are therefore unable to respond to new information being produced and shared by other team members during a mission. At MTX, ScanEagle aircraft flew optimal ISR patterns to search the island's road network for enemy forces, but could not adapt these pre-planned trajectories to better support friendly forces actions on the ground. Clearly, this capability is needed in order to fully realize unmanned systems' potential as nodes in a responsive mobile network. We propose new research that will augment existing motion planning algorithms in three ways: 1) Update underlying probabilistic models in response to detection events or external cueing/tasking received from other network nodes; 2) Explicitly incorporate network communications constraints to maintain connectivity with other network nodes; and 3) Adapt current MATLAB-based algorithms for near real-time implementation on actual vehicle autopilots. One promising approach uses Bezier curves for efficiently computing feasible vehicle trajectories in collaborative, multi-agent applications requiring spatial and temporal de-confliction under time-varying communications constraints. These improvements will make unmanned vehicles much more responsive while operating within the network control system paradigm being explored through CRUSER's MTX program.
dc.rightsThis publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.
dc.titleReal-Time Optimal Motion Planning for Responsive Mobile Networks
dc.title.alternativeReal-Time Optimal Motion Planning for Responsive Mobile Networks
dc.typePresentation
dc.contributor.departmentMechanical and Aerospace Enginerring (MAE)


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record