UAV Planning via Traversability Graphs and Time Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The planning and scheduling of multiple unmanned aerial vehicles (UAVs) in obstacle-rich areas is complex due to resource allocation, time scheduling, geometric and spatial constraints, and inter-vehicle conflict resolution, making existing optimization methods like MINLP inefficient for fast-response situations.
Innovation Solution
A method involving the discretization of obstacle fields into traversability graphs, replacing unmaneuverable sections with arcs, and determining traverse time parameters to plan and schedule UAVs efficiently, using a central controller with a discretization module and planning/scheduling module to direct vehicles in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If MINLP optimization problems are used for planning and scheduling of vehicles with vehicle dynamics, object-avoidance and conflict-resolution constraints, then comprehensive solution coverage is achieved, but computational complexity and response time increase significantly
Solution Approach 1:
The patent segments the planning and scheduling problem into two distinct parts: (1) path planning that considers vehicle dynamics and obstacle avoidance, and (2) task scheduling that assigns tasks to vehicles. This segmentation allows each sub-problem to be solved independently with appropriate algorithms, avoiding the computational burden of solving them as a single MINLP problem while maintaining comprehensive solution coverage.
Solution Approach 2:
The patent introduces an intermediary data structure called a traversability graph that serves as a bridge between vehicle dynamics constraints and task scheduling. The graph encodes feasible vehicle paths and maneuvers, allowing the scheduling algorithm to work with pre-processed feasibility information rather than directly solving complex dynamic constraints, thus reducing computational complexity.
2Reliability
If detailed vehicle dynamics constraints are incorporated into planning and scheduling, then dynamic feasibility is improved, but problem complexity and computational burden increase
Solution Approach 1:
The patent performs preliminary processing of vehicle dynamics constraints by pre-computing feasible paths and maneuvers between locations and encoding them in the traversability graph before the scheduling phase. This preliminary action separates the complex dynamics analysis from the scheduling optimization, reducing the complexity of the main planning problem while ensuring dynamic feasibility is maintained.
Data Source
AI summary
A method of implementing a plurality of unmanned vehicles over an obstacle field. The method includes obtaining a physical map of the obstacle field. Discretizing the physical map into traversable edges that avoid the obstacles, the traversable edges meeting at nodes. Replacing sections of the traversable edges that are beyond the maneuverability of the unmanned vehicles with traversable arcs. Determining traverse time parameters associated with paths through the obstacle fields, each path made up of select traversable edges and select traversable arcs and using the traverse time parameters in planning and scheduling the vehicles.