Multi-Vehicle Route Prediction Using Collision-Avoidance Hypotheses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional route prediction systems fail to accurately predict routes for multiple surrounding vehicles simultaneously due to the lack of consideration for mutual movements, leading to contradictions and increased calculation load.
Innovation Solution
A route prediction system that includes an observation unit, a signal processing unit to generate hypotheses based on collision avoidance models for surrounding vehicles, and a display unit to show predicted routes and confidence levels, reducing calculation load by focusing on feasible route combinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the degree of interference for a predicted route of a surrounding vehicle is calculated considering all predicted routes of the other surrounding vehicles, then the accuracy of route prediction is improved, but the calculation load becomes enormous
Solution Approach 1:
The patent segments the route prediction process by dividing surrounding vehicles into different groups based on their spatial relationships and collision risks. Instead of calculating interference for all vehicles simultaneously, the system performs calculations in stages: first predicting routes for vehicles with high collision risk, then progressively adding other vehicles. This segmentation reduces the computational complexity from O(n!) to a manageable sequence of smaller calculations.
Solution Approach 2:
The system performs preliminary route prediction for surrounding vehicles before calculating interference degrees. By pre-calculating predicted routes using individual vehicle models and storing them, the system avoids redundant calculations when computing interference degrees. This preliminary action prepares the data structure needed for efficient interference calculation without repeating the entire prediction process.
2Power
If routes of surrounding vehicles are predicted individually, then the calculation load is reduced, but mutual movements of plural surrounding vehicles are not considered leading to contradictions
Solution Approach 1:
The patent implements a feedback mechanism where the predicted routes of surrounding vehicles are iteratively refined. After initial individual predictions, the system calculates interference degrees and uses this feedback to adjust the predicted routes. This iterative process continues until convergence, ensuring that mutual movements are considered while maintaining computational efficiency through the feedback loop rather than exhaustive calculation.
Solution Approach 2:
The system dynamically adjusts the route prediction process based on the detected collision risks and spatial relationships. Rather than using a static individual prediction approach, the system modifies prediction parameters and recalculates routes for vehicles involved in potential conflicts. This dynamic adjustment ensures reliability by considering mutual movements only when necessary, balancing accuracy with computational load.
3Device complexity
If predetermined control rules are used to calculate the optimum predicted route, then the calculation process is simplified, but the predicted route may be erroneous when surrounding vehicles do not follow these rules
Solution Approach 1:
The patent changes the approach from using fixed predetermined control rules to using learned parameters from historical data. The system incorporates statistical parameters such as typical avoidance maneuvers, reaction times, and preference probabilities that are derived from observing actual vehicle behaviors. These parameters are continuously updated based on new data, allowing the prediction model to adapt to real-world vehicle behaviors rather than relying on rigid predetermined rules.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The route prediction system according to the invention includes a measurement unit to measure an area including a host vehicle and other moving vehicles, a vehicle detection unit to detect the host vehicle and at least two of the surrounding vehicles having collision possibilities on the basis of observation results observed by the observation unit, a hypothesis generation unit to generate plural hypotheses for the at least two of the surrounding vehicles detected by the vehicle detection unit to avoid collision, a likelihood calculation unit to calculate a likelihood indicating probability of occurrence of each of the plural hypotheses generated by the hypothesis generation unit, and a predicted route analysis unit to analyze, on the basis of the likelihood calculated by the likelihood calculation unit, predicted routes of the at least two of the surrounding vehicles, and output the analysis result. With such a configuration, in a case where plural vehicles may collide in future, predicted routes of the plural surrounding vehicles can be calculated without contradiction, improving performance on predicting the routes of the surrounding vehicles.