The field crop canopy contains a wealth of information that can provide important data for the study of crop phenotype and crop row perception methods. At present, the information collection process usually faces problems such as uneven crop height change and large differences in crop performance in different growth periods, which causes great difficulties to the adaptability and operation efficiency of ground acquisition equipment. Therefore, a distributed electric wheeled platform was designed with adjustable wheel base and ground clearance, manual remote control and automatic navigation dual-mode control. Firstly, the agronomic traits of typical field crops were analyzed, the boundary parameters of the platform structure were determined, and the whole structure design was carried out. The chassis control system was developed, a motion control system with STM32F103 chip as the core controller was established, which can support multiple steering modes such as front-wheel Ackerman steering, four-wheel Ackerman steering and in-place steering. A navigation control system with Xavier as the decision controller and GNSS/INS integrated navigation and positioning was constructed. The chassis motion performance test showed that the straight-line alignment performance of the built platform was stable and the steering was flexible, and the average lateral deviation was 0.019 m, the standard deviation was 0.017 m, the average heading deviation was 1.67 ° and standard deviation was 1.03 °. For the steering performance test, the average steering accuracy of the front wheel Ackerman was 96.57%, the average steering accuracy of the four- wheel Ackerman was 96.64%, and the average deviation of in-situ steering was 0.034 m. Taking maize as a representative crop, the crop canopy image acquisition experiment was carried out to verify the effectiveness of the built platform, and the results showed that the built platform could automatically track the planned route for row driving, ground head reversal, and cross-row alignment, and complete the non-destructive collection of canopy images. The average lateral deviation of the opposite driving was not more than 0.052 m, the standard deviation was not more than 0.029 m, the average value of the course deviation was not more than 5.653 °, and the standard deviation was not more than 3.843 °.