Path Generator for Mobile Objects Using Phantom Potential
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The artificial potential method for mobile object navigation often converges to a local minimum, preventing the generation of a movement path to the ultimate destination, and the Laplace potential method is computationally inefficient, making rapid path planning and guidance challenging.
Innovation Solution
A path generator that calculates a composite potential based on attractive and repulsive potentials, includes a local minimum determination unit to identify and avoid local minima by generating a phantom potential, and uses tag information to rapidly search for a movement path within a limited search region.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the artificial potential method is used for path generation, then the path can be automatically and flexibly generated in an object space, but the search may converge to a local minimum and fail to reach the ultimate destination
Solution Approach 1:
The patent introduces a local minimum determination unit as an intermediary component that detects when the path search has converged to a local minimum. This unit monitors the potential field and identifies convergence points, then triggers the phantom potential generation mechanism to resolve the local minimum and continue the search toward the global minimum at the destination.
Solution Approach 2:
The patent dynamically changes the potential field parameters by generating a phantom potential when a local minimum is detected. This phantom potential modifies the original potential field by adding an additional potential component that creates a new gradient direction, allowing the search to escape from the local minimum and continue toward the destination.
2Reliability
If the Laplace potential method is used to avoid local minimum, then a single minimum point is achieved, but calculation time becomes significantly long
Solution Approach 1:
The patent segments the path search process into distinct phases: normal potential-based search, local minimum detection, and phantom potential resolution. Instead of using the computationally intensive Laplace method for the entire search, the system uses the simpler artificial potential method for most of the search and only activates the phantom potential mechanism when a local minimum is detected, significantly reducing overall calculation time.
Solution Approach 2:
The patent applies the computationally heavier phantom potential method only partially - specifically when and where local minima are detected - rather than applying it universally throughout the entire search space. This selective application maintains reliability in avoiding local minimums while minimizing the loss of time associated with extensive calculations.
Data Source
AI summary
A path generator includes a map generator for generating a map for a movement space based on information on the position and shape of a mobile object, an obstacle, and a target object. It also includes a composite potential generator for calculating an attractive potential and a repulsive potential based on relative positional relationship among the mobile object, the obstacle, and the destination position, and generating a composite potential that is a sum of the attractive potential and the repulsive potential. A local minimum determination unit performs a path search in the map based on the composite potential and determines whether a convergence position of the path search is a local minimum. If it is, a phantom potential generator generates a phantom potential to be added to the potential of the convergence position. If the convergence position is the destination position, a path generator generates a movement path for the mobile object based on the result of the path search.