The Design And Verification Of The Key Modules For Multi-rotor UAV Intelligent Control Chip

In recent years, due to the advantages on flexible operation, small volume, simple structure, vertical takeoff and landing, the multi-rotor unmanned aerial vehicle(UAV) has been widely used in various fields, such as ground mapping, fighting fire in forest, road inspection and aerial photography. At present, there are many kinds of multi-rotor UAV on the market. However, most of the products apply the combination of ARM core chip and software program to control flight. As a result, these kinds of multi-rotor UAV have a disadvantage of speed for data acquisition and processing. The power consumption of chip also increases rapidly because of the low executive efficiency. What`s more, it could decrease the duration of flight.In order to solve such problems, our laboratory set up a team to design the ASIC chip for controlling UAV. In this dissertation, the key modules related to attitude reference system, GPS orientation system and motor-controlling system are designed for communication between host and peripherals, filtering and calibration of the sensors, controlling the peripherals intelligently, which are based on the analysis of multi-rotor UAV system.The module for processing the Attitude reference system data: through analyzing the output data from gyroscope, accelerometer, magnetometer, it is concluded that the output data of the three sensors include random-drift, and the output data of gyroscope and magnetometer has larger zero-drift. According to the requirement of attitude calculation, it is presented that reducing the random-drift of three sensors by smoothing filter, removing the zero drift of gyroscope by mean method, calibrating the magnetometer by ellipsoid fitting; the algorithms of refreshing automatically, serial processing and synchronous output. It designs the circuit for all the algorithms and the interfaces circuit for fundamental communication, such as I2 C, SPI serial interfaces. Through the board level verification on the 7Z020 ZYNQ, it is concluded that the speed of data acquisition and processing is improved by 51.6% compared with the way of software.The module for decoding the GPS orientation system data: first of all, it designs the UART interface for the communication of GPS module. Then the data from GPS module is divided into three categories. It presents the five-stage algorithm for decoding NMEA protocol and the algorithm for checking synchronously. At last, the circuit for the algorithms is designed and it is concluded that decoding by hardware is much faster than the way of software mode.The module for controlling the motor and remote-controller system: it acquires the PWM signal from the RC receiver and analyzes the noise which is doped in the signal. Then it designs the interface circuit of RC receiver according to the requirement for PID algorithm. Based on the flight mode of multi-rotor UAV, the channel signals are encoded into three modes for controlling the fligh t, which includes PID mode, locking mode and debugging mode. According to the flight mode of UAV and the conversion relationship between the PID data and the data of the electron speed regulator, it designs the circuit for controlling the electron speed regulator. At last, it establishes the verification and debugging platform for motor-controlling system on the board of 7Z020 ZYNQ, whose results can be observed on line by Vivado and Chipscope.