| In recent years,in the field of autonomous driving,traditional car companies and technology companies have entered the field of autonomous driving.How to obtain road information,vehicle information,and how to integrate target information for autonomous vehicles is an unavoidable issue in the field of autonomous driving.Millimeter wave radar is a full-time,all-weather vehicle radar sensor.It has the characteristics of short wavelength,bandwidth,narrow beam,and strong anti-weather interference.Therefore,millimeter-wave radar has become an indispensable environmental sensor for intelligent driving of automobiles,and has broad application prospects.In the measured data of radar equipment,it is often found that there is a large amount of clutter interference when the actual vehicle-mounted millimeter-wave radar is working,which affects the radar’s recognition of important targets(such as people,vehicles,and guardrails).Based on the radar chip AWR1642,this paper analyzes and studies the background clutter characteristics,radar signal processing technology and radar clutter suppression technology of the vehicle-mounted radar.Firstly,the scattering models of rough surfaces are studied,including Kirchhoff model.Analyzed the relevant theories of the background clutter of the vehicle-mounted millimeter-wave radar,and the background clutter environment in the work of the vehicle-mounted millimeter-wave radar.The power spectrum and amplitude spectrum of the background clutter are analyzed,and the clutter is simulated by the zero-memory nonlinear transformation method,which lays the foundation for the following target extraction.Secondly,the characteristics of MIMO radar are analyzed,and a radar chip with MIMO system is selected as the development platform based on its advantages.Analyzed the signal processing methods of the radar range,velocity,and angle dimensions under the frequency-modulated continuous wave system.And the Doppler phase compensation is performed on each channel based on the characteristics of the MIMO radar.By using a digital delay canceller,the clutter is suppressed.Then,the constant false alarm detection technology of vehicle-mounted millimeter wave radar is studied,and the technical difficulties in constant false alarm detection are analyzed,as well as the advantages and disadvantages of common constant false alarm detectors.Based on the working requirements of the vehicle-mounted millimeter wave radar,a combination optimization scheme of the constant false alarm algorithm is proposed.Finally,a road test of the vehicle-mounted millimeter-wave radar was carried out.The radar data was visualized through the host computer,and the DBSCAN algorithm was improved based on the echo characteristics of the vehicle-mounted millimeter-wave radar,which solved the problem of target recognition fragmentation.Through actual road tests,the reliability and correctness of the algorithm are further verified. |
