Autonomous Vehicle Supervision Using Weighted Controller Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous motor vehicle control systems are inadequate in managing critical situations due to limited adaptability to dynamic road conditions and interactions, such as traffic and pedestrians, which can lead to discrepancies between the reference and real trajectories.
Innovation Solution
A supervision method and device that utilize switching coefficients to weight and interpolate between different basic functions of controllers for longitudinal, lateral, and suspension operations, allowing for real-time adaptation and management of critical movements based on input data from perception, navigation, and location data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a reference trajectory definition device is used to control autonomous motor vehicle movement, then the vehicle can follow a predefined route, but discrepancies may arise between the reference trajectory and the real trajectory due to environmental constraints and vehicle dynamics
Solution Approach 1:
The control system dynamically adjusts the reference trajectory in real-time based on actual vehicle dynamics and environmental constraints. The system continuously monitors vehicle state (acceleration, steering angle, speed) and modifies the reference trajectory to account for these dynamic factors, ensuring the trajectory remains achievable and accurate under varying conditions.
Solution Approach 2:
The system implements a feedback mechanism where the actual vehicle trajectory is continuously compared with the reference trajectory. Based on the detected deviations and their causes (environmental constraints, vehicle dynamics), the reference trajectory definition device adjusts future trajectory points to prevent recurring discrepancies, creating a closed-loop control system that improves tracking accuracy over time.
2Reliability
If multiple controllers are used to manage different supervision types (longitudinal, lateral, suspension), then critical situations can be better managed, but the device complexity increases
Solution Approach 1:
The control system is segmented into three independent but coordinated controllers, each responsible for a specific supervision type: longitudinal control for speed and acceleration, lateral control for steering and position, and suspension control for vehicle stability. This segmentation allows each controller to specialize in its domain while working together through a common reference trajectory framework, managing complexity through functional decomposition.
Solution Approach 2:
The reference trajectory definition device serves as a universal coordinator that integrates inputs from all three controllers and generates a unified reference trajectory that satisfies all supervision requirements simultaneously. This multi-functional component harmonizes the specialized controllers, allowing them to work together without requiring complex inter-controller communication protocols.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The invention relates to a supervisory method for controlling an autonomous vehicle (10) according to at least two types of supervision. This method includes a step of generating switching coefficients (α1, ..., αN; α'1, ..., α'N; α"1, ..., α"N) from a general control decision. A further step involves generating a specific command (DP1, DP2, DP3) from basic functions (35221, ..., 3522N) of the considered type of supervision. The basic functions (35221, ..., 3522N) are weighted by all or part of the switching coefficients (α1, ..., αN). Each basic function (35221, ..., 3522N) represents a controller specifically designed for a particular operation of the autonomous vehicle for the considered type of supervision.