Split UAV Autopilot Control for Latency-Tolerant Flight Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional UAV autopilot systems face limitations in processing computationally intensive algorithms, leading to reduced accuracy and stability due to limited processing capabilities, latency issues, and increased power consumption, particularly when using single microcontrollers, dual processors, or non-real-time microprocessors.
Innovation Solution
A split control system configuration for UAV autopilot architecture, featuring a real-time low-level microcontroller as the main processor and a non-real-time high-level microprocessor as a co-processor, where the co-processor computes complex algorithms and desired body rate values, which are then fed to the main processor for motor control signals, maintaining stability through a rate damping loop algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single microcontroller is used for UAV autopilot system, then device complexity is reduced, but processing capability for computationally intensive algorithms is limited
Solution Approach 1:
The system divides processing tasks between two processors: a microcontroller handles real-time control tasks while a microprocessor handles computationally intensive navigation algorithms. This segmentation allows each processor to specialize in specific task types, improving overall processing capability without significantly increasing system complexity.
2Productivity
If a combination of microcontroller and microprocessor is used, then processing capability is improved, but ability to handle complexity of estimation and control logics remains limited
Solution Approach 1:
The system dynamically allocates tasks between processors based on real-time requirements. The microcontroller executes time-critical control loops with fixed priorities, while the microprocessor handles variable-complexity navigation algorithms. This dynamic task distribution allows the system to adapt to different operational complexities while maintaining real-time performance.
3Adaptability or versatility
If a single microprocessor is used, then adaptability for running various algorithms is improved, but latency is introduced compromising flight control
Solution Approach 1:
The system segments control functions into real-time critical tasks executed by the microcontroller and non-critical computation tasks executed by the microprocessor. This segmentation ensures that time-sensitive control loops maintain deterministic timing while the more adaptable microprocessor handles diverse algorithms without introducing latency to flight control.
4Productivity
If dual processors are used, then processing capability and algorithm diversity are improved, but power consumption increases affecting flight time
Solution Approach 1:
The system employs partial action by activating only the necessary processor for each specific task. The microcontroller runs continuously for real-time control, while the microprocessor is activated only when computationally intensive navigation algorithms are required. This selective activation reduces overall power consumption while maintaining the capability to execute diverse algorithms when needed.
Data Source
AI summary
A split control system for UAV incorporating auto pilot is disclosed. Control system comprises a real-time low-level main processor, and a non-real-time high-level co-processor. The co-processor computes desired body rate values and feeds them to the main processor which may be with latency. Main processor computes one or more motor control signals based on the desired body rate values. The main processor also executes a rate damping loop algorithm based on instantaneous body rate values to generate one or more motor control signals to maintain stability of the UAV even in events of latency in desired body rate values from the co-processor. Instantaneous body rate values are either obtained directly from sensors without any latency or obtained by main processor indirectly with negligible latency. Main processor acts as an intermediate between sensors and co-processor by collecting raw sensor data and feeding the data to co-processor.