In-Vehicle Route Generation Using Pre-Computed Candidate Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing in-vehicle route generation systems fail to effectively generate routes that avoid obstacles not previously encountered by the host vehicle, limiting their adaptability and safety in dynamic environments.
Innovation Solution
An in-vehicle processing device that includes a storage unit for previous route data, a candidate point generator to create alternative points based on current vehicle and environmental information, a route generator to form candidate routes, and a selector to choose the optimal route considering distance, curvature, and vehicle position, thereby avoiding new obstacles in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a route is generated based on previously traveled routes only, then the route generation is simple and fast, but the system cannot avoid new obstacles that have not previously existed
Solution Approach 1:
The system pre-generates multiple candidate routes in advance by creating candidate points along the stored route and connecting them with candidate lines. These candidate routes are prepared beforehand and stored, so when an obstacle is detected, the system can quickly select from pre-computed alternatives rather than generating routes in real-time, thus maintaining simplicity while enabling obstacle avoidance
Solution Approach 2:
The route generation system transitions from a static approach (using only stored routes) to a dynamic approach where multiple candidate routes are generated and can be selected based on current obstacle conditions. The system dynamically adjusts which candidate route to follow by evaluating obstacle positions against pre-generated alternatives, enabling adaptability without complex real-time computation
2Reliability
If multiple candidate routes are generated to avoid new obstacles, then the obstacle avoidance capability is improved, but the processing time and computational load increase
Solution Approach 1:
Multiple candidate routes are generated in advance during periods when no obstacles are present, and these pre-computed routes are stored for quick retrieval. When an obstacle is detected, the system simply evaluates which pre-generated candidate route avoids the obstacle rather than computing new routes in real-time, significantly reducing processing time while maintaining obstacle avoidance capability
Solution Approach 2:
The system generates candidate points and candidate lines at specific locations along the stored route where alternative paths might be needed. Rather than generating completely new routes throughout the entire path, candidate routes are locally adjusted in segments, reducing computational load while providing sufficient alternatives for obstacle avoidance
Data Source
AI summary
An in-vehicle processing device includes a storage part configured to store route information on routes on which a host vehicle has previously traveled; a candidate point generation part configured to generate a plurality of candidate points in a predetermined zone relative to a stored route of the route information stored in the storage part based on current peripheral information around the host vehicle and current information on the host vehicle; a candidate route generation part configured to generate a plurality of candidate routes by connecting the plurality of candidate points generated by the candidate point generation part; and a route selection part configured to select one of the plurality of candidate routes generated by the candidate route generation part.