This invention discloses an automatic material unloading method and
system for vehicles in low-profile, obstructed environments, relating to the field of industrial
automation control technology. The method includes the following steps: S1,
system initialization, creating a
Modbus continuous address data block
space object, and starting communication services,
heartbeat threads, communication detection threads, and weighbridge weighing upload threads; S2, obtaining the vehicle's scheduled unloading quantity through
license plate information, and obtaining the vehicle's arrival
signal and
tare weight information; S3, determining the unloading mode and dynamically calculating unloading parameters based on the scheduled unloading quantity and
tare weight information. This invention addresses the technical shortcomings of existing
radar monitoring solutions where material overflow occurs due to vehicles parking crookedly or to one side. It employs
deep learning visual segmentation technology to identify the boundary between the
raw material and the inner wall of the vehicle compartment in real time, accurately determining the material level status by calculating the distance between their edges. This is unaffected by vehicle parking position deviations, effectively preventing material overflow and simultaneously solving the problem of limited
radar installation in low-profile environments.