Steering Wheel Control via Obstacle Map Error Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parking support systems face challenges in accurately recognizing valid parking divisions due to vehicle position estimation errors caused by variables such as vehicle specification and road surface conditions, leading to reduced accuracy and increased parking alignment errors.
Innovation Solution
A method and apparatus that utilize an around view monitor (AVM) system and distance sensing sensors to detect parking lines and intersecting points from composite images, compensate for vehicle motion errors by calculating rotating and moving amounts, and construct an obstacle map to improve parking division recognition, thereby enhancing steering wheel control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If differential odometry scheme using rear wheel sensor data is used for vehicle position estimation, then the parking support system can be implemented with low cost, but the accuracy of recognizing valid parking division space is reduced due to vehicle specification variables and road surface state
Solution Approach 1:
The patent introduces an obstacle map as an intermediary data structure that mediates between the low-precision differential odometry and the parking division recognition task. The obstacle map integrates sensor data and image processing results to compensate for odometry errors, thereby maintaining low implementation cost while improving recognition accuracy through error compensation mechanisms
Solution Approach 2:
The system implements feedback by continuously updating the obstacle map with new sensor data and comparing detected parking lines with the stored map. This feedback loop allows the system to correct cumulative odometry errors over time and maintain accurate parking division recognition despite the limitations of differential odometry
2Device complexity
If vehicle position estimation is performed using differential odometry, then the system complexity is reduced, but the parking alignment error is increased due to systematic errors from vehicle specification mismatches and road surface conditions
Solution Approach 1:
The obstacle map serves as an intermediary that decouples the simple differential odometry from the reliability-critical parking alignment task. By using the obstacle map to store and reference parking line information, the system maintains low device complexity while improving reliability through error compensation and verification mechanisms
Solution Approach 2:
The system performs preliminary action by pre-storing parking line information in the obstacle map before the actual parking maneuver. This allows the system to have reference data available for comparison and error correction during the parking process, thereby improving alignment accuracy without increasing the complexity of the real-time position estimation system
Data Source
AI summary
Provided are a method for controlling a steering wheel and an apparatus therefore; for example, a method using an AVM composite image and a distance sensing sensor based parking support system. For example, a method for controlling a steering wheel, includes: detecting a parking line and at least two intersecting points intersecting the parking line from a first image obtained by composing images input from a plurality of cameras equipped in a vehicle; detecting a rotating amount and a moving amount of the vehicle from the first image and a second image continued to the first image; compensating for a rotating angle error and a moving error of the vehicle from the detected rotating amount and moving amount; constructing an obstacle map by reflecting the compensated rotating angle error and moving error; and controlling the steering wheel based on the constructed obstacle map.