Fleet Trajectory Planning by Dependency-Based Vehicle Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The coordination of multiple vehicles in confined sites with mutually exclusive zones is computationally complex and inefficient, particularly due to the NP-hard nature of combinatorial decisions, leading to significant computational effort in realistic scenarios.
Innovation Solution
A method involving graph decomposition and partitioning vehicles into subsets based on a dependency metric, allowing independent trajectory planning for each subset, which decomposes the nonlinear program into smaller subproblems, enabling parallel computation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a general-purpose solver is used to compute coordinated vehicle trajectories, then the solution provides optimal vehicle trajectories, but the computational complexity scales cubically with the number of vehicles and mission length
Solution Approach 1:
The patent divides the fleet of vehicles into multiple subsets based on a dependency metric that measures interaction strength between vehicles. Each subset is then processed independently through separate nonlinear programs, transforming one large cubic-complexity problem into multiple smaller problems with reduced computational burden while maintaining trajectory optimality for each subset.
2Adaptability or versatility
If the number of vehicles and MUTEX zones increases, then the coordination problem becomes more comprehensive, but the solution space grows rapidly making computation intractable
Solution Approach 1:
The patent segments the vehicle fleet into subsets based on spatial and temporal dependencies, allowing the coordination problem to be solved for each subset independently. This segmentation prevents the solution space from growing exponentially with the total number of vehicles, as each subset problem involves only a fraction of the total vehicles and MUTEX zones.
Solution Approach 2:
The patent introduces a dependency metric parameter that dynamically determines how vehicles are grouped into subsets. By adjusting this parameter, the system can balance between coordination comprehensiveness and computational tractability, adapting to different fleet sizes and mission complexities without requiring complete re-computation.
3Reliability
If trajectory planning is performed for all vehicles jointly, then safety constraints are fully enforced, but the computational effort remains significant even for realistic problem instances
Solution Approach 1:
The patent segments vehicles into subsets based on their interaction dependencies, ensuring that safety constraints are fully enforced within each subset while allowing parallel computation. This segmentation maintains reliability by preserving all necessary safety constraints for interacting vehicles while improving productivity through independent parallel solution of subset problems.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method of planning trajectories for vehicles operating in a common environment, wherein movements of each vehicle are controllable by a control signal, comprising: for each vehicle, obtaining (1312) a predefined vehicle path to be traversed; performing a first computation (1314) to obtain a vehicle crossing order O at each mutually exclusive - MUTEX - zone between two vehicle paths, wherein the first computation is subject to safety constraints; and performing trajectory planning (1320) subject to the obtained vehicle crossing order at the MUTEX zones, to obtain a control signal for each of the vehicles, the method further comprising assigning (1316) a dependency metric to each pair of vehicles, and partitioning (1318) the vehicles into a number NSG of vehicle subsets such that the dependency metric exceeds a threshold within each, wherein the trajectory planning is performed as multiple independent subproblems (1320-1, ..., 1320-NSG), each relating to one of the vehicle subsets.