Extensible distributed multi-redundancy unmanned aerial vehicle intelligent flight control computer

Abstract

The invention discloses an extensible distributed redundancy unmanned aerial vehicle intelligent flight control computer, which is characterized in that a distributed system adopts a hardware similar redundancy form; the system is composed of a redundancy flight control module node taking an XCZU7EV device as a carrier, a redundancy comprehensive information processing intelligent calculation module node taking the XCZU7EV device as a carrier, and a group of dual-channel FlexRay redundancy time trigger buses. A time-sharing partition mapping strategy based on task division is adopted, multiple tasks are efficiently mapped, intelligent processing highly related to a control law and a decision is distributed to a flight control module, and intelligent processing related to intelligent perception is distributed to a redundancy comprehensive information processing intelligent calculation module. An open and comprehensive information processing comprehensive design method is provided and realized, integrated comprehensive information processing of flight control, navigation calculation and atmosphere data calculation is realized, and a fault-tolerant design method based on dual-core self-monitoring is provided and applied to prevent Byzantine faults.

Description

technical field [0001] The invention belongs to the technical field of computers, and in particular relates to an intelligent flight control computer for unmanned aerial vehicles. Background technique [0002] In the field of aeronautical computers, redundant and fault-tolerant design techniques are generally used to improve the reliability and fault-tolerant capabilities of flight control systems, and the more common ones are double-redundant computers and triple-redundant computers. In the event-triggered flight control system, the current public flight control computer is usually based on a limited number of nodes for node fault-tolerant design, which has poor scalability. At the same time, due to the lack of support for intelligent computing and limited bandwidth restrictions, it cannot support neural networks. Large data calculation and transmission processing of intelligent computing nodes; in the time-triggered flight control system, on the one hand, it lacks elastic ...

AI Technical Summary

Problems solved by technology

In the event-triggered flight control system, the current public flight control computer is usually based on a limited number of nodes for node fault-tolerant design, which has poor scalability. At the same time, due to the lack of support for intelligent computing and limited bandwidth restrictions, it cannot support neural networks. Large data calculation and transmission processing of intelligent computing nodes; in the time-triggered flight control system, on the one hand, it lacks elastic scalability and fault-tolerant architecture support, and on the other hand, it lacks hardware architecture support for intelligent flight control, so it cannot Adapt to the increasingly mature neural network flight control information processing calculation

Images