Lane Keep Assist Control Circuit With Neural PID Gain Tuning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing PID controllers in automotive vehicle lane keep assist systems face suboptimal tuning due to internal and external changes in vehicle operating conditions, leading to inefficiencies in maintaining lane position and heading angle control.
Innovation Solution
A lane keep assist system incorporating a neural network to dynamically tune the P, I, and D gain terms of a PID controller using environmental variables, allowing continuous adaptation to changing conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If fixed gain terms are used in PID controller during design phase, then device complexity is reduced, but control accuracy deteriorates under changing operating conditions
Solution Approach 1:
The patent applies dynamics by transitioning from fixed gain terms to dynamically adjustable gain terms. The neural network continuously updates the P, I, and D gain values based on real-time operating conditions such as vehicle speed, steering angle, and lane curvature, enabling the PID controller to adapt to changing conditions and maintain optimal control accuracy without increasing fundamental system complexity
Solution Approach 2:
The patent replaces the traditional mechanical approach of manually tuning fixed gain terms with an intelligent system using neural networks. This substitution allows automatic, real-time adjustment of controller parameters based on environmental variables, eliminating the need for complex manual recalibration while improving control precision across varying operating conditions
2Ease of operation
If fixed gain terms are used in PID controller, then ease of operation is improved, but adaptability to changing conditions deteriorates
Solution Approach 1:
The patent implements self-service by enabling the PID controller to automatically adjust its own gain terms through the neural network without requiring manual intervention. The system monitors operating conditions and autonomously recalibrates the P, I, and D values, maintaining ease of operation while achieving high adaptability to changing vehicle speed, steering conditions, and lane geometry
Solution Approach 2:
The patent applies parameter changes by dynamically modifying the gain terms (Kp, Ki, Kd) of the PID controller based on environmental variables. The neural network processes inputs such as vehicle speed, steering angle, and lane curvature to continuously update controller parameters, enabling the system to adapt to varying operating conditions while maintaining simple operation through automated adjustment
3Measurement precision
If neural network is added to dynamically tune gain terms, then control accuracy is improved, but device complexity increases
Solution Approach 1:
The patent applies universality by designing the neural network to handle multiple control functions simultaneously. The same neural network structure tunes all three gain terms (P, I, and D) and processes various environmental variables (vehicle speed, steering angle, lane curvature), reducing the need for separate control mechanisms while achieving comprehensive adaptability and maintaining manageable system complexity
Solution Approach 2:
The patent applies preliminary action by pre-training the neural network with extensive data covering various operating conditions before deployment. This pre-training enables the network to make accurate gain term adjustments from the outset, reducing the complexity of real-time computation and allowing the system to achieve high control accuracy without requiring overly complex online processing capabilities
4Adaptability or versatility
If neural network dynamically tunes gain terms, then adaptability to environmental conditions is improved, but loss of time for computation increases
Solution Approach 1:
The patent applies periodic action by updating the gain terms at strategically selected intervals rather than continuously. The neural network recalibrates the P, I, and D values at key moments such as significant changes in vehicle speed, steering angle, or lane curvature, reducing computational overhead while maintaining real-time adaptability to changing operating conditions
Solution Approach 2:
The patent applies preliminary action by pre-computing and storing optimal gain term values for various operating conditions during the training phase. This pre-computation allows the neural network to quickly retrieve and apply appropriate gain values during operation, minimizing real-time computation time while maintaining high adaptability to environmental conditions
Data Source
AI summary
A lane keep assist system for an automotive vehicle includes an electric power steering assembly that is responsive to an output of the control system, the motor applying a torque to a part of a steering gear to steer the vehicle along a highway. The lane keep assist system assists a driver in keeping the vehicle in a lane of a highway, in which the control circuit comprises A PID Controller which receives at an input a target lane position for the closed-loop control system and provides as an output a control signal for a motor of the electric power steering assembly. The controller is arranged in a closed loop with the motor configured to minimise an error value indicative of the difference between the target lane position and the actual lane position of the vehicle.