Autopilot Adaptive Control for UAS Platform Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current unmanned aircraft systems (UAS) face limitations in performing novel tasks due to the computational constraints of their autopilot systems, requiring laborious tuning of traditional PID controller parameters and lacking airframe-independent control algorithms, which restricts their ability to operate effectively in various environments and transfer controllers between different platforms efficiently.
Innovation Solution
The development of a Rapid Controller Transfer (RCT) method using a nonparametric, Bayesian adaptive control algorithm that enables an airframe-independent autopilot system, specifically designed for fixed-wing UAS, allowing for the transfer of autopilot hardware with minimal performance impact from a well-known to a poorly understood system, leveraging Gaussian Processes for online learning and uncertainty modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional PID controller parameters are used for each UAV platform, then control reliability is maintained, but development time and cost increase significantly due to extensive manual tuning required for each platform
Solution Approach 1:
The controller performs self-tuning by automatically adapting its parameters based on real-time feedback from the UAV platform. The system uses iterative learning algorithms to adjust control parameters without requiring manual intervention, enabling the controller to optimize its performance autonomously for each specific platform while maintaining reliability.
Solution Approach 2:
The invention dynamically changes controller parameters based on the specific characteristics of each UAV platform. Instead of using fixed PID parameters, the system adapts parameters such as proportional gains, integral gains, and derivative gains according to the platform's unique dynamics, thereby eliminating the need for extensive manual tuning while maintaining control reliability.
2Reliability
If platform-specific control tuning is performed for each UAV, then control performance is optimized, but the process becomes extremely costly and time consuming
Solution Approach 1:
The controller is designed with universal adaptability to work across multiple UAV platforms without requiring platform-specific customization. The system uses a standardized interface and adaptive algorithms that automatically adjust to different platform characteristics, enabling a single controller design to serve multiple functions across various UAV types while maintaining optimal performance.
Solution Approach 2:
The controller autonomously adapts to each platform's specific characteristics through self-tuning algorithms, eliminating the need for expensive manual tuning processes. The system automatically identifies platform parameters and adjusts control gains accordingly, significantly reducing development costs while maintaining optimized control performance.
3Ease of manufacture
If COTS autopilots with low price point are used, then cost is reduced, but hardware quality and software reliability deteriorate
Solution Approach 1:
The open-source autopilot incorporates self-diagnostic and self-tuning capabilities that automatically monitor system health and adjust parameters in real-time. This autonomous adaptation compensates for potential hardware limitations and software imperfections, maintaining high reliability without requiring expensive proprietary components.
Solution Approach 2:
The system implements comprehensive feedback mechanisms that continuously monitor control performance and system state. This real-time feedback enables the controller to detect and correct errors, adapt to changing conditions, and maintain reliable operation even with cost-effective COTS hardware and open-source software, thereby achieving high reliability without compromising on cost efficiency.
4Reliability
If open-source autopilots with high reliability are used, then control reliability is improved, but capability to perform novel tasks is limited and price increases
Solution Approach 1:
The autopilot implements dynamic adaptability through real-time parameter adjustment and online learning capabilities. The system can dynamically modify control strategies based on the specific task requirements and platform characteristics, enabling it to perform novel tasks such as STOL, deep stall landings, and acrobatic maneuvers while maintaining the high reliability of open-source software without being constrained by proprietary limitations.
Data Source
AI summary
According to an embodiment, there is provided an onboard integrated computational system for an unmanned aircraft system (“Stabilis” autopilot). This is an integrated suite of hardware, software, and data-to-decisions services that are designed to meet the needs of business and research developers of UAS. Stabilis is designed to accelerate the development of any UAS platform and avionics system; it does so with hardware modularity and software adaptation. The Stabilis offers multiple technological advantages technological advantages including: Plug-and-adapt functionality; Data-to-decisions capability; and, On board parallelization capability.