Autonomous Vehicle Controller Coefficient Tuning via PSO Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion planning and control for autonomous driving vehicles do not accurately account for differences in vehicle types, leading to suboptimal controller coefficients that deviate from optimal points due to deteriorating vehicle parameters, lacking efficient tuning methods.
Innovation Solution
The method involves dividing driving parameters into ranges, using particle swarm optimization to determine optimal controller coefficients, and creating mapping tables to configure controllers in real-time, iteratively determining local and global best coefficients for precise control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional motion planning and control methods are used with fixed controller coefficients, then the system is simple to implement, but the control accuracy deteriorates as vehicle parameters change
Solution Approach 1:
The patent implements dynamic adaptation of controller coefficients by dividing driving parameters into multiple ranges and selecting different coefficient sets based on current operating conditions. The system dynamically switches between pre-calculated coefficient sets corresponding to different vehicle states (e.g., speed ranges, acceleration levels), allowing the controller to adapt to changing vehicle parameters without requiring real-time recalculation, thus maintaining control accuracy while managing complexity.
Solution Approach 2:
The patent performs preliminary calculation and storage of optimal controller coefficients for multiple driving parameter ranges before actual operation. By pre-computing coefficient sets for different vehicle states and storing them in lookup tables, the system eliminates the need for complex real-time optimization during driving, reducing computational burden while ensuring optimal control accuracy is achieved for each operating condition.
2Measurement precision
If controller coefficients are tuned manually for different driving parameters, then control precision can be improved, but the tuning process becomes time-consuming and complex
Solution Approach 1:
The patent performs preliminary optimization of controller coefficients offline before deployment. A computational algorithm automatically calculates optimal coefficient sets for multiple driving parameter ranges and stores them in pre-computed lookup tables. During actual autonomous driving operation, the system simply retrieves the appropriate coefficient set based on current driving parameters, eliminating time-consuming manual tuning processes while maintaining high precision control.
Solution Approach 2:
The patent creates multiple copies of controller coefficient sets, each optimized for specific driving parameter ranges. Instead of manually adjusting a single coefficient set, the system maintains parallel copies of coefficients tailored to different operating conditions (e.g., low speed, high speed, acceleration, deceleration). This allows instant switching between optimized coefficient copies based on current state, achieving precision without repeated tuning time investment.
3Adaptability or versatility
If the same motion planning and control is applied to all vehicle types, then the system is universally applicable, but the control performance deteriorates for specific vehicle characteristics
Solution Approach 1:
The patent applies local quality by tailoring controller coefficients to specific vehicle characteristics and operating conditions. Instead of using uniform control parameters for all vehicles, the system divides driving parameters into localized ranges and assigns specific coefficient sets optimized for each range. This allows the controller to adapt its characteristics to match the specific vehicle type and current state, achieving both universality through the framework and precision through localized optimization.
Solution Approach 2:
The patent changes controller parameters dynamically based on driving conditions and vehicle characteristics. By dividing driving parameters (speed, acceleration, heading) into multiple ranges and associating different coefficient sets with each range, the system automatically adjusts control parameters to match the specific vehicle type and operating context. This parameter adaptation mechanism maintains universal applicability while achieving vehicle-specific control performance.
Data Source
AI summary
Driving parameters (e.g., speed, heading direction) that an autonomous driving vehicle (ADV) likely utilize as target driving parameters are grouped into a number of ranges and one of the driving parameters in each range is selected as a driving parameter representative or a target driving parameter representing the respective range or segment. For each of the target driving parameters representing the ranges, a particle swarm optimization method is utilized to derive a set of most optimized coefficients for a controller (e.g., speed controller, steering controller) for controlling an ADV. A driving parameter to coefficient (parameter/coefficient) mapping table is generated to map a particular driving parameter representing a range of driving parameter to a set of one or more coefficients of a particular controller. The parameter/coefficient mapping table is utilized at real-time to configure a controller in response to a particular target driving parameter using the corresponding coefficients.