Model Reference Adaptive Controller for Vehicle Actuation Dynamics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicle control systems face challenges with time-latency and actuation dynamic delays in control subsystems such as brakes, throttle, and steering, leading to unsmooth responses during rapid maneuvers, due to the oversimplified structure of PID controllers which cannot fully address these issues.
Innovation Solution
The implementation of a model reference adaptive controller (MRAC) that determines a reference actuation output based on subsystem dynamics, calculates an adaptive gain from errors, and iteratively adjusts actuation commands to minimize delays, allowing the system to adaptively compensate for time-latency and dynamic delays without needing to know the specific causes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a PID controller is used to control the actuation system, then the control structure remains simple and easy to implement, but the system cannot fully address actuation time-latency and dynamic delay problems
Solution Approach 1:
The patent applies dynamics by transitioning from a static PID controller to a dynamic model reference adaptive controller (MRAC) that continuously adapts its parameters based on real-time system performance. The MRAC uses a reference model to define desired dynamic characteristics and adjusts controller gains dynamically to match the actual system response, enabling the system to handle time-varying actuation delays and nonlinearities effectively
Solution Approach 2:
The patent implements feedback by introducing an adaptive feedback mechanism where the MRAC continuously monitors the difference between the actual actuator response and the reference model output. This feedback drives the adaptation of controller parameters, allowing the system to compensate for time-latency and dynamic delays by adjusting control actions based on historical performance data and real-time error signals
2Productivity
If conventional control commands are issued to control subsystems, then the control system operates with minimal processing, but time-latency of 50-100 milliseconds occurs due to data collection and processing
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing optimal control commands and their expected responses in a database during system operation. When a control command is needed, the MRAC queries this pre-computed data to quickly determine appropriate control actions, significantly reducing the real-time processing time and latency compared to computing control commands from scratch during operation
3Speed
If actuation commands are issued rapidly during sharp turns or lane changes, then the vehicle responds quickly to maneuver commands, but actuation dynamic delay of several hundreds of milliseconds causes unsmooth and uncomfortable responses
Solution Approach 1:
The patent applies universality by creating a unified MRAC framework that handles multiple control subsystems (steering, braking, acceleration) through a single adaptive control architecture. This universal controller uses the same reference model and adaptation mechanism across different actuation systems, enabling coordinated and smooth responses during complex maneuvers while maintaining the ability to handle rapid commands across all subsystems
Data Source
AI summary
Systems and methods are disclosed for reducing second order dynamics delays in a control subsystem (e.g. throttle, braking, or steering) in an autonomous driving vehicle (ADV). A control input is received from an ADV perception and planning system. The control input is translated in a control command to a control subsystem of the ADV. A reference actuation output is obtained from a storage of the ADV. The reference actuation output is a smoothed output that accounts for second order actuation dynamic delays attributable to the control subsystem actuator. Based on a difference between the control input and the reference actuation output, adaptive gains are determined and applied to the input control signal to reduce error between the control output and the reference actuation output.