Dynamic Dither Noise Filtering in Electric Power Steering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dither noise in Electric Power Steering (EPS) systems, primarily caused by handwheel torque sensors, leads to discomfort for operators due to inherent noise in analog signals and quantization noise in digital sensors, necessitating a reduction or elimination of this low-frequency noise.
Innovation Solution
A method and apparatus for attenuating dither noise through dynamic filtering of motor torque commands using filter parameters computed based on multiple signals such as motor velocity, acceleration, and handwheel torque, with an arbitration module determining final filter parameters to configure the filter effectively, thereby reducing noise while maintaining system stability and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If dither noise filtering is applied to reduce noise, then operator comfort is improved, but system responsiveness may be degraded
Solution Approach 1:
The filter parameters are made dynamic rather than static. The arbitration module selects different filter parameters based on real-time operating conditions such as vehicle speed, steering angle, and steering torque. This allows the system to adapt the filtering strength to current operational context, reducing noise when appropriate while maintaining responsiveness when needed.
Solution Approach 2:
The system changes filter parameters (such as cutoff frequency, filter order) based on operating conditions. By arbitrating between multiple filter parameters and selecting the most appropriate one for current conditions, the system optimizes the balance between noise reduction and responsiveness without being locked into a single fixed filter configuration.
2Object-affected harmful factors
If multiple filter parameters are computed and arbitrated, then filtering effectiveness is improved, but device complexity increases
Solution Approach 1:
The filter parameter determination is segmented into multiple independent parameters that can be computed separately based on different operating conditions. The arbitration module then selects the appropriate combination of parameters, breaking down the complex task of filter design into manageable segments that can be independently optimized and combined.
Solution Approach 2:
The system uses its own operating parameters (vehicle speed, steering angle, torque) to automatically determine the appropriate filter settings. The arbitration module self-adjusts the filter parameters based on real-time sensor data without requiring external calibration or manual intervention, making the complexity management self-contained.
3Object-affected harmful factors
If dynamic filter configuration is implemented, then noise reduction performance is improved, but calibration difficulty increases
Solution Approach 1:
Multiple filter parameters are pre-computed and stored based on different operating conditions during system design. The arbitration module simply needs to select from these pre-computed parameters based on current conditions, rather than requiring real-time complex calculations or calibration. This preliminary preparation simplifies the calibration process significantly.
Solution Approach 2:
The arbitration module serves multiple functions: it selects filter parameters, adapts to different operating conditions, and optimizes noise reduction across various scenarios. This multi-functional approach consolidates what would otherwise require multiple separate calibration processes into a single unified selection mechanism.
Data Source
AI summary
Technical solutions are described for attenuating dither noise in a steering system. An example method includes computing multiple filter parameters, each filter parameter based on a corresponding signal in the steering system. The method further includes determining at least one final filter parameter from the plurality of filter parameters by arbitrating the plurality of filter parameters. The method further includes dynamically configuring a filter using the at least one final filter parameter. Further, the method includes filtering a motor torque command using the filter, a filtered motor torque command being applied to a motor to generate a corresponding amount of torque.


