Linear Path Planning for Vehicle Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current path-planning approaches for navigating vehicles, such as Dijkstra's method and A* path planning, require excessive processing due to calculating cost values for each cell in the cost map, leading to inefficient obstacle avoidance and increased computational demands, especially in small portable systems.
Innovation Solution
A method that identifies a current linear path between a vehicle's current location and a target location, evaluates cells for obstacles, and instructs the vehicle to travel along a different path if an obstacle is detected, using the extent of the obstacle and potential locations to select an intermediate location with the lowest cost, thereby reducing unnecessary path calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional path-planning approaches (Dijkstra's method, A* path planning) are used to calculate cost values for each cell in the cost map, then obstacle avoidance is achieved, but processing requirements increase proportionally to the square of the linear dimensions of the region of interest
Solution Approach 1:
The patent segments the path planning process into discrete linear path segments between waypoints rather than evaluating every cell in the cost map. This segmentation reduces the computational domain from the entire region of interest to only the relevant path segments, thereby reducing processing requirements while maintaining obstacle avoidance capability.
Solution Approach 2:
The patent extracts only the essential path information needed for navigation by identifying linear paths between waypoints and evaluating obstacles only along these paths. This extraction eliminates unnecessary cost map cell evaluations, significantly reducing processing requirements while preserving the core obstacle avoidance function.
2Reliability
If traditional path-planning approaches proactively check and determine path costs through cost map cells, then complete path evaluation is achieved, but the vehicle may back up unnecessarily even though that course is not selected
Solution Approach 1:
The patent performs preliminary evaluation of linear path segments between waypoints to identify obstacles before the vehicle commits to movement. By proactively checking only the relevant linear path segments rather than all cost map cells, the system determines viable paths in advance, preventing unnecessary backup movements while maintaining complete path evaluation.
Solution Approach 2:
The patent applies partial action by evaluating only the necessary linear path segments between waypoints rather than performing exhaustive evaluation of all cost map cells. This partial evaluation is sufficient for safe navigation and prevents unnecessary vehicle maneuvers while reducing computational overhead.
3Reliability
If the processing required for path planning increases according to the total area of the cost map, then comprehensive obstacle detection is achieved, but the ability of small portable systems to accommodate the processing is exceeded
Solution Approach 1:
The patent segments the computational task to evaluate only linear path segments between waypoints rather than the entire cost map area. This segmentation maintains comprehensive obstacle detection along the relevant paths while reducing the total processing area, enabling implementation on small portable systems with limited computational resources.
Solution Approach 2:
The patent extracts and evaluates only the essential linear path information needed for navigation, removing the need to process the entire cost map area. This extraction achieves comprehensive obstacle detection along the vehicle's path while reducing computational requirements to levels suitable for portable systems.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A solution for navigating a vehicle is provided. The vehicle is navigated from a start location to a target location using a set of linear paths. A current linear path is identified between a current location of the vehicle and the target location. At least a portion of the current linear path is evaluated for a presence of an obstacle. In response to no obstacle being present along the current linear path, the vehicle is instructed to travel along the current linear path to the target location. In response to an obstacle being present, the vehicle is instructed to travel along a different linear path from the current location to an intermediate location, which is selected based on an extent of the obstacle and the current linear path. The process can be repeated until the vehicle arrives at the target location.