Steady-State Lateral Deviation Compensation in Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-driving vehicles experience steady-state lateral deviation due to inconsistencies in wheel positioning and IMU angular deviation, leading to safety and stability issues, particularly during turns.
Innovation Solution
A method and device that determine if a vehicle is traveling on a straight road by detecting road curvature, steering wheel angle, and velocity, and collect lateral deviation values to identify and compensate for steady-state lateral deviation in real-time, using a detection module, collection module, determination module, and compensation module.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continuous deviation rectification method is used to eliminate lateral deviation, then lateral deviation is reduced, but the vehicle produces leftward and rightward swaying and reduces self-driving stability
Solution Approach 1:
The patent applies periodic action by implementing deviation rectification only under specific periodic conditions (straight road, low curvature, steady state) rather than continuous rectification. The system periodically checks whether the vehicle meets the predetermined conditions and applies correction only when appropriate, thereby eliminating lateral deviation while avoiding continuous swaying and maintaining self-driving stability.
2Reliability
If deviation rectification is applied continuously, then lateral deviation is corrected, but self-driving safety and stability deteriorate due to swaying
Solution Approach 1:
The patent applies local quality by making the deviation rectification effect conditional and localized rather than uniform and continuous. The system identifies specific local conditions (straight road sections with low curvature and steady state) and applies correction only in those localized contexts, thereby improving lateral deviation correction while avoiding the harmful swaying effects that would compromise safety and stability.
Solution Approach 2:
The patent applies preliminary anti-action by preemptively identifying and eliminating the root causes of lateral deviation (wheel positioning inconsistency, IMU angular deviation) before they can cause safety issues. The system proactively detects and corrects steady-state lateral deviation under controlled conditions, preventing the accumulation of errors that would lead to instability and safety problems.
3Device complexity
If steady-state lateral deviation is not compensated, then system complexity remains low, but self-driving accuracy and safety deteriorate
Solution Approach 1:
The patent applies parameter changes by modifying the lateral deviation parameter under specific conditions. The system changes the lateral deviation parameter from its original uncorrected value to a compensated value when the vehicle is in steady state on a straight road with low curvature, thereby improving self-driving accuracy while maintaining relatively low system complexity through conditional parameter adjustment.
Data Source
AI summary
A method and a device for eliminating a steady-state lateral deviation and a storage medium are provided. The method includes: determining whether a vehicle travels on a straight road; collecting a lateral deviation value of the vehicle in a case that the vehicle travels on a straight road; determining whether the vehicle has the steady-state lateral deviation based on the collected lateral deviation value; and compensating the steady-state lateral deviation of the vehicle in real time based on the collected lateral deviation value in a case that the vehicle has the steady-state lateral deviation. The vehicle has the steady-state lateral deviation may run steadily, a direction of the vehicle is more accurate, self-driving safety and efficiency are improved, the sense of leftward and rightward swaying brought by a continuous deviation rectification method in the conventional art is effectively eliminated, and self-driving stability is improved.