Nonlinear Vehicle Control via Hardware Accelerator
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous control systems for vehicles like UAVs face limitations in computation time and real-time performance, especially when dealing with external disturbances, due to the use of general-purpose processors, which can lead to errors and instability.
Innovation Solution
A configurable hardware accelerator is introduced, independent of the general-purpose processor, that implements a nonlinear control system based on a generic model of translational and rotational dynamics, allowing for high-frequency, real-time operations and robust control, with a pipeline architecture optimizing computational speed and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If general-purpose processors are used for autonomous control, then system flexibility and ease of operation are improved, but computation time increases and real-time performance deteriorates
Solution Approach 1:
The control system is segmented into two independent parts: a general-purpose processor for high-level autonomous control functions and a dedicated hardware accelerator for real-time control computations. This segmentation allows each component to specialize in its optimal function, with the hardware accelerator handling time-critical calculations at fixed frequencies while the general-purpose processor manages flexible decision-making.
Solution Approach 2:
A communication interface acts as an intermediary between the general-purpose processor and the hardware accelerator. This mediator transfers control parameters and sensor data between the two systems, enabling coordinated operation where the general-purpose processor can update control parameters without interfering with the real-time computation frequency of the hardware accelerator.
2Adaptability or versatility
If general-purpose processors are used for autonomous control, then adaptability is improved, but reliability under external disturbances deteriorates
Solution Approach 1:
The system separates adaptive control parameter management (handled by the flexible general-purpose processor) from stable real-time execution (handled by the deterministic hardware accelerator). This segmentation ensures that adaptability updates do not compromise the stability of core control loops.
Solution Approach 2:
The hardware accelerator is designed with a fixed-frequency architecture that provides inherent protection against external disturbances. By pre-establishing a deterministic computation schedule and isolating real-time control from system interruptions, the system cushions against potential instability before disturbances can propagate.
3Measurement precision
If high-frequency real-time operations are implemented, then control accuracy is improved, but device complexity increases
Solution Approach 1:
The system replaces complex software-based real-time control with a dedicated hardware accelerator that performs control computations in hardware. This substitution achieves high-frequency real-time operations with deterministic timing, eliminating the need for complex scheduling and interrupt management that would otherwise be required in a software-only approach.
4Loss of time
If isolated hardware architecture is used, then real-time performance is improved, but ease of operation deteriorates
Solution Approach 1:
A communication interface serves as an intermediary that simplifies integration between the isolated hardware accelerator and the general-purpose processor. This mediator handles data transfer and coordination, allowing the hardware accelerator to maintain its isolated, interruption-free architecture while still receiving control parameters and sending results to the general-purpose processor for high-level decision-making.
Data Source
AI summary
Methods and apparatus to implement nonlinear control of vehicles moved using multiple motors, apparatus, systems and articles of manufacture are disclosed. An example apparatus includes a logic circuit configured to calculate virtual position control variables for a vehicle moved using multiple motors. The virtual position control variables calculated based on an application of a control law to position information of the vehicle. The control law is derived from a nonlinear model of movement of the vehicle. The logic circuit is configured further to calculate control inputs based on the virtual position control variables. The control inputs are to control the motors to navigate the vehicle along a designated path of movement.


