Design of UAV flight control system based on SmartFusion

0 Introduction
The flight control system is an important part of the UAV and is the operating platform of the flight control algorithm. Its performance is directly related to whether the UAV can fly safely and reliably. With the development of aviation technology, the UAV flight control system is developing in the direction of multi-function, high precision, miniaturization and reusability. High precision requires the UAV control system to have high precision, good stability and be able to adapt to complex external environments. Therefore, the control algorithm is relatively complex, with fast calculation speed and high precision; miniaturization puts higher requirements on the weight and volume of the control system, requiring the performance of the control system to be as high as possible and the volume to be as small as possible. In addition, the UAV flight control system must also have the characteristics of real-time, reliability, low cost and low power consumption. Based on the above considerations, this paper designs a UAV flight control system based on SmartFusion from the perspective of practical engineering applications.

1 Overall design of flight control system
The flight control system has two main functions on UAVs: one is flight control, that is, the UAV maintains the stability of the aircraft attitude and track in the air, and changes the aircraft attitude and track according to the ground radio remote control command or the pre-set altitude, route, heading, attitude angle, etc., to ensure the stable flight of the aircraft, which is usually called autopilot; the other is flight management, that is, completing the flight status parameter collection, navigation calculation, telemetry data transmission, fault diagnosis and processing, emergency situation processing, and control and management of mission equipment.
The flight control system mainly completes three functional tasks, and its hierarchy is composed of three layers: the lowest task is to improve the inherent damping of UAV movement and gust mitigation - the damper function in the three axis directions; the second layer's task is to stabilize the attitude angle of the UAV - the function of the basic autopilot (mainly for angular motion control); the third layer's task is to control the flight altitude, track and flight speed to achieve a higher level of autopilot function. The principle block diagram of the flight control system is shown in Figure 1.

From the above analysis, it is easy to know that the flight control system is mainly composed of three parts: flight controller, sensor (or sensitive element), and servo. The basic architecture of the UAV flight control system is shown in Figure 2.

The flight controller is a control computer with SmartFusion as the core. It is the central control unit of the drone, responsible for the coordination of various units on the aircraft and data transmission between the ground station. At the same time, according to the control algorithm and the command of the ground station, the aircraft is kept flying in a certain attitude.
Sensors include electronic compass, angular rate sensor, altitude sensor, accelerometer and GPS receiver. The electronic compass uses Honeywell HMR3300, which can measure heading, tilt and roll angle, and the output mode is SPI serial output. The angular rate sensor uses ADXRS300 from ADI, and its output voltage is proportional to the yaw rate. The altitude sensor uses the MPX4115A atmospheric pressure sensor manufactured using the piezoresistive effect principle of semiconductor materials. The accelerometer uses ADXL203 from ADI, which is a capacitive accelerometer with high precision, high stability and low power consumption. The output voltage is proportional to the acceleration value after signal conditioning. The GPS receiver is GPS-G03A(H), which is an ultra-low power GPS antenna receiver. It can provide information such as longitude and latitude, time and speed, and transmit data in the form of RS 232 serial port.
The servo includes elevator, rudder and aileron. The control signal of the servo is a pulse width modulation signal, which is convenient for interface with the flight control computer.

2 Flight control computer hardware circuit design
The flight control computer of the drone is the core of the drone flight control system, and its hardware structure is shown in Figure 3. Its functions are mainly realized by the SmartFusion single-chip minimum system and peripheral circuits. In addition, there is a power module to provide various voltages required by the system and manage the power supply battery.

The SmartFusion series combines logic, microcontroller subsystem (MSS) and programmable analog modules, namely Actel's proven FPGA architecture and ARM Cortex-M3 hard-core processor subsystem, as well as programmable FLASH analog modules. It implements an easy-to-use, fully customizable system design platform, allowing embedded designers to quickly optimize hardware/software and make trade-offs without making board-level changes. In SmartFusion devices, all data is transferred from the processor to the FPGA, or from the analog module to the processor, or between the FPGA and the on-chip analog module. In addition, Actel's FLASHLock technology also provides excellent IP security. [page]
The main features and functions of SmartFusion are reflected in:
(1) Fully functional FPGA. SmartFusion devices have Actel's proven ProASIC3 FPGA architecture based on fast flash technology, using advanced 130 nm seven-layer flash CMOS process technology, system gate density ranges from 60K to 500K, and has an operating frequency of 350 MHz and a maximum of 204 I/Os. This combination can integrate existing functions from other devices, significantly reducing board space and overall system power consumption.
(2) Microcontroller subsystem. The intelligence of the device is added to the FPGA in the form of a microcontroller subsystem with an ARM Cortex-M3 processor hard core running at 100 MHz, all standard peripherals and functions including: multi-layer AHB communication matrix, throughput up to 16 Gb/s, 10M/100M Ethernet MAC and SPI with RMI interface, I2C, UART and 32-bit timer. With up to 512 KB flash, 64 KB SRAM and external memory controller (EMC) and 8-channel DMA controller.
(3) Programmable analog module. The innovative proprietary analog computing engine (ACE) can perform sampling sorting and calculation, and can share the analog initialization and processing tasks of the ARM Cortex-M3 processor. The programmable analog includes: ADC and DAC with an accuracy of 1%, up to 3 12-bit ADCs with a sampling frequency of 600KS/s, up to 3 12-bit first-order DACs, 10 50 ns high-speed comparators, and integrated multiple temperature, voltage and current monitoring functions.
Here, the A2F200M3 of the SmartFusion series is selected as the core chip. The main function of its MSS is to maintain communication with the ground station, collect attitude angle data and GPS positioning data, send control commands to FPGA, and run related flight control algorithms and navigation control algorithms. The main function of the FPGA architecture is to collect and measure sensor data, receive MSS commands and drive servos. FPGA shares some of the tasks that could originally be implemented by MSS, allowing MSS to have more time to run algorithms to improve the overall performance of the system.
The peripheral interface circuit is mainly composed of RS 232 interface, GPS interface, SD card interface, sensor signal acquisition channel, engine speed measurement channel, battery voltage detection channel, etc. GPS and PC serial port connections both require a MAX232 chip for level conversion. The SD card is used to store flight parameters and image data. Data can be transmitted between MSS and FPGA through GPIO. The functions of servo drive, wireless receiver signal detection, signal control, etc. are all realized by the internal logic circuit of the chip and are written in VerilogHDL language. In view of the requirements of flight control and navigation accuracy, the module uses TI's A/D chip TLC3548 to collect the voltage signals output by the three-axis accelerometer, altitude sensor, and triangular rate gyro. The A/D and drive component block diagram is shown in Figure 4.

3 Software Design
Due to the complexity of the programs run by the flight control computer, the large amount of information, and the high requirements for real-time performance and stability, the programming method using a single-task sequential mechanism can no longer meet the requirements of the flight control system. Therefore, the flight control software uses a real-time embedded operating system μC/OS-Ⅱ. μC/OS-Ⅱ is a real-time operating system designed specifically for embedded applications of computers. It is a scalable, preemptive multi-task real-time kernel based on static priority. It is highly portable and particularly suitable for microprocessors and microcontrollers. Its real-time performance and kernel robustness have been proven in a large number of practical applications.

The application program of the flight control system is divided into initialization module, data acquisition module, control solution module, attitude reading module, GPS receiving module, telemetry sending module, and control quantity output module. The relationship between tasks and functional module resources is shown in Figure 5. When the flight control system starts running, after completing the self-check, the flight control computer first performs the initial alignment of the inertial navigation system and the binding of task parameters, receives the initial attitude and position information sent by the initial alignment device, and then waits for the start command of the control system. After the flight control system is started, the initialization settings are performed, and the corresponding program modules are called according to the task priority to complete the scheduled tasks. The common data between tasks are coordinated in the form of shared variables, but certain protection measures need to be taken.

4 Conclusion
The UAV flight control system based on SmartFusion has the advantages of small size, high precision, fast computing speed and scalability. In particular, the software uses the μC/OS-Ⅱ real-time operating system, which makes the flight control system have real-time characteristics. The flight control system has been applied to a certain type of UAV, and the flight verification shows that the system design meets the requirements. The flight control system designed in this paper has certain engineering application value and provides a reference for the design of the flight control system.