| High-precision synchronous acquisition of multi-sensor data is one of the key technologies for autonomous driving.With the support and promotion of advanced technologies around the world,autonomous driving has become a hot spot,where numerous.Companies and universities attach great importance.For example,large foreign technology corporations,such as Google and Uber,have launched self-driving car projects since 2005.In addition,Baidu,Tucson and other domestic top technology companies have launched the same projects from 2013.However,the current multi-sensor data acquisition systems for automatic driving are mostly integrated by multiple modules,which cause notable drawbacks including high overall cost,low time synchronization accuracy,large volume and relatively cumbersome assembly.In this thesis,a high-integration multi-sensor data synchronous acquisition controller using FPGA as the core processing chip is designed for the above problems.To achieve high integration,high time precision,multi-sensor data acquisition and control of synchronous controller,the principle contributions of the thesis are as follows:(1)Establish a high-precision clock reference.The high-stability quartz crystal oscillator is used as the clock source of the synchronizer,and it is calibrated by combining the GPS PPS pulse and NEMA data to make full use of the long-term stability of the PPS pulse of GPS and the high stability of the high-stability quartz crystal.A high precision time reference within the measurement time range.(2)Synchronization of multiple sensors.According to the characteristics of each sensor,the sensors such as the inertial navigation unit and the camera are actively synchronized.The three-dimensional laser radar adopts timing synchronization,and the wheel encoder adopts passive synchronization.The sensor data is collected in pure hardware and the precise time label is used as the synchronous alignment mark.,to achieve high-precision synchronization of multi-sensor raw data.(3)Complete the schematic diagram and PCB design,drawing and debugging of the overall hardware circuit of the synchronous controller.Designed multi-sensor data acquisition circuits such as inertial navigation unit,wheel encoder,differential GPS,3D laser radar,etc.;high-speed interface and storage circuit such as Gigabit network,USB3.0,USB2.0,mSATA,TF card,and the whole hardware circuit Debugging enables acquisition,transmission and storage of multi-sensor data.(4)Completing the writing and debugging of the FPGA program for each sensor control and data synchronous acquisition.The FPGA chip is used as the carrier and combined with the external hardware circuit to establish a high-precision time reference through the hardware description language.The SPI interface is designed to control and collect the inertial navigation data.The UART interface is designed to synchronize the time of the 3D laser radar to realize the interaction between the FPGA and the TX2 command.Synchronous data transmission of encoder and camera;the control program of CY7C68013A chip is designed to transfer parallel data to USB serial data,realizing high-speed transmission of a large number of sensor raw data between FPGA and TX2.The synchronous controller is installed on a vehicle for the automatic driving test.The results reveal that the synchronous controller can provide multi-sensor high-precision synchronous raw data such as GPS,3D laser radar,Inertial navigation unit,wheel encoder and camera for automatic driving in real time.The automatic driving positioning accuracy of the synchronous controller can reach ±10cm,and the synchronous controller system can achieve absolute time accuracy of ±1μs.The synchronous controller also has the advantages of small size,light weight,low cost,high integration,fast data processing speed and strong expandability,etc.,and has been successfully applied in the automatic park of Shenzhen Zhihui Technology Co.,Ltd.and the Chinese Academy of Agricultural Sciences.In driving projects,it also has broad application prospects in many fields such as power inspection,high-precision positioning,navigation and mobile measurement. |