Steering Torque Sensor Kalman Filter Friction Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining a driver's manual torque at a steering wheel are hindered by frictional interactions and mechanical irregularities in steering columns, making it difficult to accurately assess driver activity, especially in hands-free travel scenarios.
Innovation Solution
A method and device utilizing a Kalman filter to estimate driver's manual torque by accounting for frictional torque, sensed through a steering torque sensor, and incorporating a physical model of the steering system to improve accuracy and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a steering torque sensor is used to measure torque at the steering column, then torque measurement is achieved, but frictional torque and mechanical irregularities reduce measurement precision
Solution Approach 1:
The patent replaces direct mechanical torque measurement with a computational estimation approach using a Kalman filter. Instead of relying solely on the mechanical steering torque sensor output, the system uses a mathematical model that incorporates steering angle speed, steering torque, and frictional torque characteristics to estimate the driver's manual torque, thereby eliminating the direct impact of mechanical friction and irregularities on measurement precision
Solution Approach 2:
The patent introduces a Kalman filter as an intermediary computational layer between the steering torque sensor and the final torque measurement. This filter processes the raw sensor data along with steering angle speed and frictional torque models to produce a corrected torque estimate, effectively mediating the harmful effects of friction and mechanical irregularities
2Reliability
If frictional torque is present in the steering column, then mechanical connections function, but driver activity detection reliability deteriorates
Solution Approach 1:
The patent implements a feedback mechanism through the Kalman filter that continuously monitors steering angle speed and steering torque, compares them with the frictional torque model, and adjusts the torque estimation accordingly. This feedback loop enables the system to compensate for frictional torque effects in real-time, maintaining reliable driver activity detection despite the presence of friction in the mechanical system
Solution Approach 2:
The patent converts the harmful frictional torque into a beneficial factor by using it as an input parameter for the Kalman filter estimation. The frictional torque, which was previously a source of measurement error, is now explicitly modeled and used to improve the accuracy of driver's manual torque estimation, turning a harmful effect into a useful correction factor
Data Source
AI summary
A method for determining a driver's manual torque at a steering wheel of a vehicle which includes sensing a steering angle speed by a steering angle sensor, sensing a steering torque by a steering torque sensor at a steering column connected to the steering wheel, estimating a driver's manual torque applied by a driver at the steering wheel based on the sensed steering angle speed and the sensed steering torque by a Kalman filter by a controller, wherein during the estimation of the driver's manual torque a frictional torque is considered, and the frictional torque is estimated based on the steering torque which is sensed by the steering torque sensor, wherein the estimated frictional torque is taken into account as an interference factor during the estimation of the driver's manual torque in the Kalman filter. Also disclosed is an associated device.