Autonomous Vehicle Trajectory Planning for Early Lane Change Intent
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles struggle to accurately predict the intent of surrounding vehicles when they intend to change lanes, leading to potentially unsafe and unnatural driving maneuvers due to uncertainty and data noise in perception systems.
Innovation Solution
An autonomous vehicle system uses sensors and machine learning models to monitor the lateral position and velocity of nearby vehicles, predicting their trajectory and likelihood of collision, and adjusts its path to avoid potential collisions by determining alternative trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the autonomous vehicle uses basic perception systems to detect surrounding vehicles, then the system complexity is low, but the measurement precision of lateral position and velocity is insufficient leading to inaccurate intent prediction
Solution Approach 1:
The system performs preliminary actions by continuously monitoring and storing lateral position and velocity data before a lane change occurs. This historical data is used to calculate lateral acceleration and predict future trajectories, enabling accurate intent prediction before the actual lane change happens, thus improving measurement precision without requiring more complex hardware.
2Reliability
If the autonomous vehicle monitors and predicts trajectories of all surrounding vehicles, then the safety is improved, but the computational complexity and processing time increase
Solution Approach 1:
The system applies local quality by focusing computational resources on vehicles that exhibit lane change intent characteristics (specific lateral position and velocity patterns) rather than uniformly processing all surrounding vehicles. This selective approach maintains high collision avoidance reliability while reducing overall computational complexity by identifying and prioritizing only relevant targets.
3Speed
If the autonomous vehicle reacts immediately to detected lane change attempts, then the response time is reduced, but the driving maneuvers become unnatural and uncomfortable
Solution Approach 1:
The system performs preliminary prediction of lane change intent using lateral acceleration calculations and trajectory projection before the actual lane change occurs. This allows the autonomous vehicle to prepare smooth, gradual trajectory adjustments in advance, maintaining fast response while ensuring natural and comfortable driving maneuvers by avoiding sudden reactions.
Data Source
AI summary
A method comprises periodically monitoring, by a processor, lateral position and velocity of vehicle within a predetermined distance from an autonomous vehicle, the vehicle moving in a direction having at least one common attribute with the autonomous vehicle; executing, by the processor, a computer model using the monitored lateral position and velocity of the vehicle, to predict whether a trajectory for the vehicle; and when a current trajectory of the autonomous has a likelihood of collision with the predicted trajectory of the vehicle that satisfies a threshold, determining, by the processor, an alternative trajectory for the autonomous vehicle.