Unified MPC Reference Tracking Across Two Autonomous Driving Modes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems lack a unified control scheme for semi-autonomous and fully autonomous driving modes, leading to inefficiencies and increased hardware requirements due to separate control systems for each mode.
Innovation Solution
Implementing a common control scheme using reference tracking and the concept of a pseudo-driver input, which allows for the use of Model Predictive Control (MPC) to compare and refine driving commands across both modes, reducing hardware needs and improving tracking performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate control systems are used for semi-autonomous and fully autonomous driving modes, then each mode can be optimized independently, but hardware requirements increase and system complexity increases
Solution Approach 1:
The patent implements a universal control system that handles both semi-autonomous and fully autonomous driving modes through a single MPC controller. The controller receives driving inputs from either a human driver or a pseudo-driver model and generates appropriate control commands for both operating modes, eliminating the need for separate control systems and reducing hardware requirements while maintaining mode-specific optimization capabilities
2Reliability
If separate control systems are used for semi-autonomous and fully autonomous driving modes, then each mode can be optimized independently, but system complexity increases
Solution Approach 1:
The control system is designed as a multi-functional universal system that operates in both semi-autonomous and fully autonomous modes through a single MPC controller. The system accepts different input types (human driver input or pseudo-driver input) and produces appropriate control outputs for each mode, thereby reducing system complexity while preserving the ability to optimize control parameters specific to each driving mode
3Device complexity
If a common control scheme is used for both semi-autonomous and fully autonomous modes, then hardware requirements are reduced and system complexity is reduced, but tracking performance may be compromised
Solution Approach 1:
The MPC controller dynamically adapts its behavior based on the operating mode. In semi-autonomous mode, it processes human driver inputs with appropriate filtering and prediction. In fully autonomous mode, it uses pseudo-driver inputs generated by a predictive model. The controller adjusts its internal parameters and control strategy in real-time to maintain optimal tracking performance across both modes while using a single hardware platform
Solution Approach 2:
The pseudo-driver model serves as an intermediary that generates realistic driving inputs for fully autonomous mode, allowing the common control system to process inputs in a unified manner. This intermediary enables the single MPC controller to achieve tracking performance comparable to mode-specific systems by translating fully autonomous operating conditions into a format compatible with the control algorithm
Data Source
AI summary
Systems and methods of using a common control scheme to autonomously controlling a vehicle during two different autonomous driving modes are provided. In particular, embodiments of the presently disclosed technology incorporate reference tracking for driving input and vehicle state into this common control scheme. In some embodiments, this common control scheme may be implemented using Model Predictive Control (MPC).