MPC Assistive Torque Calculation for Steering Conflict Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing assistive torque calculation methods for Advanced Driver Assistance Systems (ADAS) struggle to determine optimal values for assistive torque applied to the steering column, often resulting in inconsistent or conflicting forces with the driver's torque, leading to sub-optimal guidance and reduced driver comfort.
Innovation Solution
A computer-implemented method using Model Predictive Control (MPC) to calculate assistive torque by determining optimal values over a time horizon, considering driver torque prediction and vehicle behavior, which minimizes a cost function based on desired reference states, ensuring smooth cooperation between the driver and ADAS.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If assistive torque is applied to the steering column by ADAS, then tracking performance and safety are improved, but driver comfort deteriorates due to torque conflicts and intrusive sensations
Solution Approach 1:
The system continuously measures the driver's applied torque and uses this feedback to dynamically adjust the assistive torque. The measurement unit captures real-time driver torque data, which feeds into the calculation unit to optimize the assistive torque values, ensuring they align with driver intentions and minimize conflicting sensations.
Solution Approach 2:
The system dynamically changes the parameter of assistive torque based on measured driver torque and predicted driver behavior. By adjusting the assistive torque parameter in real-time according to driver actions and predictions, the system optimizes tracking performance while maintaining driver comfort and avoiding intrusive sensations.
2Reliability
If assistive torque values are increased to improve guidance, then tracking performance is improved, but driver comfort deteriorates due to excessive or inconsistent torque application
Solution Approach 1:
The system applies partial assistive torque rather than full corrective torque, adjusting the assistance level to match driver needs and predictions. By applying only the necessary portion of torque required for guidance while considering driver intentions, the system avoids excessive torque application that would cause conflicts or discomfort.
Solution Approach 2:
The calculation unit acts as an intermediary between the driver's torque application and the ADAS assistive torque. It processes driver torque measurements and predictions to determine optimal assistive torque values, mediating the interaction to ensure smooth cooperation and minimize torque conflicts.
3Stability of the object's composition
If driver torque is measured and used for assistive torque calculation, then assistive torque consistency is improved, but system complexity increases due to additional sensors and calculation requirements
Solution Approach 1:
The measurement unit serves multiple functions: it measures driver torque for assistive torque calculation, provides feedback for real-time adjustments, and supplies data for prediction algorithms. This multi-functionality reduces the need for separate dedicated components, thereby limiting the increase in system complexity while improving assistive torque consistency.
Solution Approach 2:
The system uses the driver's own torque measurements to generate assistive torque, creating a self-service mechanism where driver input directly informs the assistance provided. This approach improves consistency by basing assistance on actual driver actions while avoiding the need for complex external prediction models.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented assistive torque calculation method for calculating an assistive torque (Tc(k)) to be applied to a steering unit of a vehicle, in addition to a driver torque (DTk). The method comprises: S12) determining optimal values (Tc(k)) of the assistive torque over a time horizon (Np) using a Model Predictive Control solver (MPC); and S14) outputting the optimal value (Tc(k)) determined for the next time point (tk+1); wherein a prediction unit of the MPC solver (MPC) comprises a driver torque prediction unit (DTP) and a vehicle behaviour prediction unit (VBP); the driver torque prediction unit (DTP) being configured to predict a predicted driver torque (Tact); the vehicle behaviour prediction unit (VBP) being configured to output an estimated state of the vehicle at the next time point (tk+1). The invention further encompasses a system, a program, a data storage medium, a vehicle, adapted to perform the assistive torque calculation method.