Automated Driving Path Planning Using Probability and Driver Behavior
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated driving vehicles face challenges in determining a path when there are many moving objects, as predicted positions of these objects spread out, making it difficult to find suitable positions for the vehicle to be present at future points in time, leading to the inability to establish a path.
Innovation Solution
A vehicle control apparatus that acquires information on the surrounding environment, calculates a first value for the probability of objects being present at future positions and a second value based on travel data from a predetermined driver, and determines a path by selecting positions where the difference between these values reaches a minimum or is below a threshold, combining these values to ensure the vehicle can navigate through areas where objects are less likely to be present.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle determines path positions by avoiding all predicted positions of moving objects, then collision avoidance is improved, but the ability to establish a path deteriorates when many objects are present
Solution Approach 1:
The patent applies local quality by differentiating between different spatial regions around the vehicle. Instead of uniformly avoiding all predicted object positions, the system identifies specific regions where objects are likely to be present and applies avoidance logic selectively. The path determination unit calculates predicted positions for multiple objects and creates localized avoidance zones, allowing the vehicle to navigate through areas where objects are less likely to be present while maintaining collision avoidance in high-probability zones.
Solution Approach 2:
The patent changes the parameter of path determination from binary avoidance (avoid/not avoid) to a probabilistic approach. By calculating predicted positions for multiple future time points and evaluating the likelihood of object presence at each position, the system transforms the path planning problem into finding positions with acceptable probability thresholds. This allows the vehicle to establish paths even when many objects are present, by selecting positions where the combined probability of object presence is acceptable.
2Reliability
If the vehicle considers only object presence probability for path determination, then collision avoidance is improved, but natural driving behavior deteriorates
Solution Approach 1:
The patent merges two different value systems: the first value representing object presence probability (safety criterion) and the second value representing natural driving behavior (comfort criterion). The path determination unit combines these values to select positions that satisfy both criteria. Specifically, it evaluates positions based on the first value to ensure safety, then refines the selection using the second value to ensure the path feels natural and comfortable, achieving a balance between collision avoidance and natural driving behavior.
Solution Approach 2:
The patent implements feedback by using travel data from predetermined drivers to inform the path determination process. The system learns from historical driving behavior and uses this information to adjust path selection. When multiple positions satisfy the safety criterion (first value), the feedback mechanism selects among them based on what a natural driver would choose (second value), creating a closed-loop system that adapts to achieve both safety and naturalness.
Data Source
AI summary
A vehicle control apparatus configured to control automated driving of a vehicle acquires information relating to a situation in a surrounding area of the vehicle, acquires, for each of a plurality of positions, a first value relating to a probability that an object that is present in the surrounding area will be present at a future point in time and a second value obtained based on travel data of a predetermined driver based on the information, and determines a path on which the vehicle is to move, by selecting positions at which the vehicle is to be present at a plurality of future points in time from the plurality of positions based on combinations of the first values and the second values.