Vehicle Motion Planning for Erratic Driving Risk Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motion planning systems for autonomous vehicles fail to account for collision risks caused by unexpected driver actions, erratic driving behaviors, and damaged vehicle components.
Innovation Solution
A system and method for motion planning that uses vehicle sensors to measure remote vehicles, determines risk scores for location cells using a machine learning model, and adjusts the vehicle's planned path based on these risk scores and the behavior class of the remote vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If motion planning systems use traditional path optimization methods, then computational efficiency is maintained, but they fail to account for unexpected driver actions and erratic driving behaviors
Solution Approach 1:
The patent introduces a machine learning model as an intermediary component between sensor data collection and motion planning. This model processes remote vehicle behavior data and generates risk scores, effectively mediating between raw sensor inputs and path optimization decisions. The intermediary handles the complexity of behavior prediction, allowing the traditional motion planning system to maintain its computational efficiency while incorporating advanced risk assessment capabilities.
2Reliability
If motion planning systems incorporate comprehensive risk assessment for all location cells, then safety against erratic behaviors is improved, but computational load and processing time increase
Solution Approach 1:
The patent applies local quality by assigning different risk scores to different location cells based on their specific characteristics and the predicted behavior of remote vehicles. Instead of uniformly processing all location cells with the same computational resources, the system identifies high-risk cells requiring detailed analysis and low-risk cells that can be processed more quickly. This selective approach maintains safety while reducing overall computational time.
3Adaptability or versatility
If motion planning systems use simple path optimization, then real-time responsiveness is maintained, but they cannot adapt to non-standard driving behaviors
Solution Approach 1:
The patent implements preliminary action by using the machine learning model to predict potential erratic behaviors and assign risk scores to location cells before the actual path optimization occurs. This pre-processing of risk information allows the traditional motion planning algorithm to work with pre-evaluated data, maintaining real-time responsiveness while incorporating adaptability to non-standard behaviors through the预先 calculated risk assessments.
Data Source
AI summary
A system for motion planning for a vehicle includes at least one vehicle sensor for determining information about an environment surrounding the vehicle and a controller in electrical communication with the at least one vehicle sensor. The controller is programmed to perform a plurality of measurements of a remote vehicle using the at least one vehicle sensor. The plurality of measurements includes at least a plurality of position measurements of the remote vehicle. The controller is further programmed to determine a risk score for each of a plurality of location cells in an environment surrounding the remote vehicle based at least in part on the plurality of measurements of the remote vehicle. The controller is further programmed to adjust a planned path of the vehicle based at least in part on the risk score of each of the plurality of location cells in the environment surrounding the remote vehicle.