Route Prediction Device Using Collision Risk Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional route prediction devices face increased computation load, inaccurate risk evaluation, and reduced accuracy when predicting future positions, especially when avoiding obstacles, and the A* algorithm-based systems fail to consider motion dynamics, leading to inefficient processing times.
Innovation Solution
A route prediction device that includes a tracking processor, collision object detector, route prediction unit, collision risk estimator, collision deciding unit, and avoidance route selector, which use collision avoidance models to estimate routes, calculate risks, and select optimal collision avoidance models to reduce computing load and ensure low collision risk.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple prediction tracks are generated to calculate existence probabilities, then route prediction accuracy is improved, but computation load increases
Solution Approach 1:
The patent extracts only the necessary prediction tracks based on collision risk assessment rather than generating all possible prediction tracks. The collision risk estimation unit selectively processes prediction tracks that have potential collision risk, eliminating unnecessary computations for tracks with no collision risk.
Solution Approach 2:
The patent performs preliminary collision risk estimation before detailed route prediction. By calculating collision risk first using simplified models, the system identifies which prediction tracks require detailed analysis, avoiding computationally intensive processing for tracks that won't result in collisions.
2Device complexity
If A* algorithm is used for route determination with lattice points, then route planning is simplified, but motion dynamics of moving bodies are not considered
Solution Approach 1:
The patent segments the route prediction process into two stages: first using A* algorithm for coarse route planning on lattice points, then refining the prediction by incorporating motion dynamics models for each identified route segment. This allows the system to benefit from the simplicity of A* while adding dynamic behavior consideration.
Solution Approach 2:
The patent introduces dynamic behavior models that adjust the prediction based on the moving body's acceleration, turning radius, and other dynamic parameters. These dynamic models are applied to the routes generated by A* algorithm, allowing the system to consider motion dynamics while maintaining the computational efficiency of lattice-based planning.
3Productivity
If linear uniform velocity assumption is used for future position calculation, then computation is simplified, but accuracy deteriorates when target changes route to avoid obstacles
Solution Approach 1:
The patent applies linear uniform velocity assumption only as a baseline prediction, then adds correction terms that account for route change behavior when obstacles are detected. The system performs partial correction by adjusting the prediction based on obstacle proximity and estimated avoidance maneuvers, rather than fully simulating complex avoidance behavior in all cases.
4Adaptability or versatility
If risk calculation depends on parameters without clear method, then flexibility is maintained, but risk evaluation accuracy is reduced
Solution Approach 1:
The patent transforms abstract risk parameters into concrete, quantifiable metrics by defining specific calculation methods for collision risk. The collision risk estimation unit uses standardized formulas that combine relative position, velocity, and predicted trajectory to produce a numerical risk value, making the evaluation both accurate and reproducible while maintaining flexibility through adjustable parameters.
Data Source
AI summary
A route prediction unit estimates a route of an object of interest with respect to a target object based on collision avoidance models. A collision risk estimation unit calculates collision risks between the object of interest and target object for each collision avoidance model. A collision deciding unit decides the presence or absence of a collision from the collision risks and feeds back a collision avoidance model correction value to the route prediction unit when it is determined that the collision occurs. A collision avoidance route selector selects any of the plurality of collision avoidance models in which the absence of collision is decided by the collision deciding unit, and selects a route of the collision avoidance model as a route for avoiding the collision between the objects. The route prediction unit performs a new route prediction using the collision avoidance model correction value.