A deep tunnel blast hole charging operation environment dynamic modeling and perception method

Abstract

The present application relates to a kind of deep tunnel blast hole charging operation environment dynamic modeling and sensing method, comprising the following steps: real-time monitoring dust concentration and illumination intensity in tunnel by sensing combination, and based on the dust concentration and illumination intensity measured, the weight of each sensor data in sensing combination is adjusted by dynamic weight distribution algorithm;Lightweight time sequence feature correlation analysis network is introduced to correlate the sensor data of continuous frame, real-time track dynamic target trajectory, generate the environment model containing dynamic target identification;Combine charging operation site working condition feedback data, dynamically correct environment model error.The present application is accurately extracted blast hole position and rock wall profile static characteristics by adaptive multi-sensing fusion with dynamic weight distribution algorithm, relies on lightweight time sequence feature correlation analysis network, combines with the device posture data of inertial measurement unit, real-time track the trajectory of dynamic target such as operating personnel, rockfall and distinguish type.

Images