Device and method for monitoring of pig attitudes based on micro-inertial sensor

Abstract

A pig posture monitoring device and method based on a micro-inertial sensor, the device includes an information collection module for collecting individual information of pigs; a posture calculation module for fusing and filtering the information collected by the information collection module and outputting the posture angle; and The PC host computer used to recognize the attitude of pigs according to the output results of the attitude calculation module, the invention saves packaging space, and adopts the fourth-order Runge-Kutta optimization algorithm to update the attitude information, through the dynamic Kalman filter model, using the acceleration The gyroscope and magnetometer compensate the drift of the gyroscope, realize the complementary advantages of the gyroscope attitude and the accelerometer and magnetometer attitude, improve the accuracy and reliability of the dynamic measurement of the system, not only significantly reduce the equipment cost, but also have strong applicability. It can meet the requirements of real-time monitoring, and can replace the breeders to observe the posture of the pigs at each moment, so as to detect the suspected pigs early and reduce economic losses.

Description

technical field [0001] The invention relates to the technical field of pig posture monitoring, in particular to a pig posture monitoring device and method based on a micro-inertial sensor. Background technique [0002] Pig breeding occupies an important position in my country's animal husbandry. In recent years, it has developed rapidly, and its output and quality have been steadily improved. The breeding model is gradually developing in the direction of intensification and factoryization. However, the overall breeding efficiency is low. The main reason is that the breeding density of farms has increased. The traditional method of relying on manual monitoring of pig health and pig behavior can no longer meet the needs of modern large-scale farming, resulting in sub-healthy animals. Discovery and early warning, and then the disease occurs frequently, the mortality rate of live pigs is high, and the quality of pork is unqualified. Animal behavior patterns are the basis for di...

AI Technical Summary

Problems solved by technology

[0006] The video information measurement method based on machine vision is currently the most widely used pig attitude measurement method. Although it can capture the real-time motion state, the workload of the later moving target extraction process is heavy, including the identification, tracking, and spatial coordinates of marker points. A single simple feature extraction algorithm is difficult to meet the universality requirements of the algorithm, and there is a camera line of sight and a limited shooting range, which is easily affected by on-site lighting.

Images