Autonomous Driving Controller Steering Override Reference Value
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In autonomous driving systems, the transition from lane keeping to override mode based on steering angle control can result in an unsmooth steering feeling for drivers, as the system fails to consider the driver's tendency, leading to a significant difference in steering experience.
Innovation Solution
An autonomous driving controller that collects driving data, calculates a steering override reference value based on the driver's information and tendency, and adjusts the control mode accordingly to minimize the difference in steering experience by varying the steering override reference value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If steering angle is used as control input for autonomous driving, then feedback control for steering angle becomes possible which is more robust to disturbance, but steering override depending on specific condition causes large sense of difference and unsmooth steering feeling
Solution Approach 1:
The patent applies dynamics by making the steering override reference value adjustable and adaptable. The system dynamically adjusts the override reference value based on learned driver tendencies, transforming a static control parameter into a dynamic one that adapts to individual driver preferences. This resolves the contradiction by maintaining robust feedback control while smoothing the transition during override modes through personalized adjustment.
Solution Approach 2:
The patent changes the parameter of steering override reference value from a fixed constant to a variable that can be adjusted based on driver tendency learning. By learning the driver's steering characteristics and adjusting the override reference value accordingly, the system maintains the robustness of steering angle control while reducing the sense of difference during mode transitions, thus improving steering smoothness.
2Measurement precision
If autonomous driving controller performs steering override based on steering angle control, then control precision is improved, but driver's unsmooth steering feeling increases due to not considering driver's tendency
Solution Approach 1:
The patent applies preliminary action by learning the driver's steering tendency in advance before actual autonomous driving operation. The system collects and analyzes driver steering data during manual driving phases, builds a tendency model, and uses this pre-learned information to adjust the steering override reference value. This preliminary learning action enables the system to provide smooth, personalized control during override modes while maintaining high precision.
Solution Approach 2:
The patent implements feedback by using the learned driver tendency information to continuously adjust the steering override reference value. The system monitors driver steering behavior, learns from this data, and feeds this knowledge back into the control algorithm to personalize the override characteristics. This feedback loop ensures that control precision is maintained while adapting to individual driver preferences for smooth steering transitions.
Data Source
AI summary
An autonomous driving controller includes: a processor to collect driving data when a vehicle is traveling and calculate a steering override reference value, which is a criterion of determining an override mode, based on the collected driving data; and a storage to store the collected driving data and a set of instructions executed by the processor to calculate the steering override reference value. In particular, the processor controls autonomous driving by varying the steering override reference value based on the collected driving data or information regarding a driver of the vehicle.