Vehicle Route Planning with Localization Uncertainty Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional algorithms for determining vehicle routes are inadequate for multi-vehicle rendezvous scenarios, as they fail to account for the localisation uncertainty of multiple vehicles, leading to potential route failures.
Innovation Solution
A path planning process using belief state planners to construct a graph within a state space, accounting for vehicle localization uncertainty, and applying a path shortening algorithm to determine routes that minimize collision risks and optimize vehicle positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional algorithms are used for route determination, then the route planning process is simple, but the reliability of multi-vehicle rendezvous is poor due to unaccounted localization uncertainty
Solution Approach 1:
The route determination process is segmented into distinct phases: graph construction within state space, path identification, and path shortening algorithm application. This segmentation allows each phase to address specific aspects of the problem (uncertainty modeling, route finding, optimization) independently, improving reliability while managing complexity through modular processing
Solution Approach 2:
The system performs preliminary actions by constructing a comprehensive graph within the state space that incorporates localization uncertainty before actual route execution. This pre-planning phase prepares multiple potential paths and evaluates them under uncertainty conditions, ensuring reliable rendezvous while the actual execution follows predetermined optimized routes
2Loss of energy
If path shortening algorithm is applied to optimize routes, then the route cost is reduced, but the computational complexity increases
Solution Approach 1:
The path shortening algorithm applies partial optimization by selectively modifying only those portions of the path that can be improved without complete re-planning. Instead of recalculating entire routes, the algorithm performs localized adjustments to reduce cost while maintaining overall path validity, achieving energy efficiency with reduced computational burden
Solution Approach 2:
The path shortening algorithm operates continuously on the identified path, iteratively applying optimizations until no further improvements are possible. This continuous refinement process gradually reduces route cost through multiple passes of local optimization, achieving significant energy savings while distributing computational load over time rather than requiring a single complex calculation
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A method and apparatus for determining a route for a vehicle (6), the method comprising: measuring a position of the vehicle (6); providing a specification of a region (12) into which the vehicle (6) is to be moved; and, using the measurements and specification, determining the vehicle route. The route determination process comprises constructing a graph (34) within a state space (X) of the vehicle (6), identifying, within the graph (34), a path for the vehicle (6), performing a path shortening algorithm on the identified path, and, using the shortened path, determining the vehicle route. The path shortening algorithm comprises: selecting two vertices along the path that are separated by at least two edges; connecting the two selected vertices with an additional edge; and, depending on certain cost values, removing the edges and vertices by which the selected vertices are connected, and including, in the path, the additional edge.