Obstacle Position Prediction Evaluation Using Probabilistic Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of automated driving control is compromised when the prediction accuracy of an obstacle's future position detected by in-vehicle sensors is low, leading to suboptimal vehicle control.
Innovation Solution
A prediction accuracy evaluation method and system that generates a predicted distribution of an obstacle's position based on initial detection information and determines prediction abnormality by comparing it with subsequent detected positions, using a processor to assess the accuracy of the predicted distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single predicted position is used for obstacle prediction, then the prediction process is simple, but the prediction accuracy is low
Solution Approach 1:
The patent transitions from predicting a single point position to predicting a distribution across multiple possible positions. Instead of outputting one predicted position, the system generates a probability distribution indicating likelihoods of the obstacle being at various positions, thereby adding a dimensional aspect of probability to the prediction and improving accuracy.
Solution Approach 2:
The patent changes the prediction output parameter from a single position coordinate to a distribution of positions with associated probabilities. This parameter transformation allows the system to express uncertainty and multiple possibilities, enhancing prediction accuracy without requiring fundamentally new prediction mechanisms.
2Measurement precision
If only the predicted position is considered, then the evaluation is simple, but the evaluation accuracy is insufficient
Solution Approach 1:
The patent introduces a feedback mechanism where the predicted distribution is compared with the actual detected position. The system evaluates whether the actual position falls within the predicted distribution and uses this feedback to determine prediction abnormality, thereby improving evaluation accuracy through iterative validation.
Solution Approach 2:
The patent replaces simple point-to-point comparison with a probabilistic distribution-based evaluation. Instead of mechanically comparing single coordinates, the system uses statistical methods to assess whether the actual position is consistent with the predicted distribution, substituting deterministic comparison with probabilistic reasoning.
Data Source
AI summary
A prediction accuracy evaluation method is executed by a computer. The prediction accuracy evaluation method includes a process of acquiring detection information. The detection information indicates a detected position, a detected velocity, an error range of the detected position, and an error range of the detected velocity of an obstacle detected by using a sensor mounted on a moving body. The prediction accuracy evaluation method further includes: a predicted distribution generation process that generates a predicted distribution of a position of the obstacle at a second time later than a first time, based on first detection information that is the detection information at the first time; and a prediction abnormality determination process that determines whether or not the predicted distribution is abnormal based on the predicted distribution and a second detected position that is the detected position of the obstacle at the second time.