4results about How to "Improve control responsiveness" patented technology

This invention provides an integrated method for pruning and controlling large-scale models of unmanned aerial vehicles (UAVs), belonging to the field of model lightweighting technology. The proposed method significantly reduces model size and computational load, greatly improves inference speed and control response performance, and enables the deployment of complex neural network controllers on resource-constrained airborne platforms. Simultaneously, the lightweight model maintains or approaches the control accuracy and stability of the original uncompressed model. This method has wide applicability and strong scalability, covering various scenarios from basic flight control to complex mission control, providing a feasible, efficient, and cost-effective solution for applying large models to airborne controllers of UAVs, and possesses high engineering practical value.

The utility model provides a laser pumping drive circuit, and relates to the technical field of drive circuits. The output end of a voltage-controlled constant current source unit is connected with a pumping laser; the output end of the dual-mode current setting unit is connected with the reference input end of the voltage-controlled constant current source unit; the output end of the temperature control driving unit is connected with the thermoelectric refrigerating unit, and the feedback input end of the temperature control driving unit is used for receiving a temperature sensing signal; the analog input end of the multi-channel signal acquisition unit is respectively connected with a current sampling point of the voltage-controlled constant current source unit, the output end of the monitoring photodiode and a temperature sensing signal node; the positive power supply output end and the negative power supply output end of the power supply management unit are respectively connected with the power supply ends of the voltage-controlled constant current source unit and the multi-channel signal acquisition unit, and the enabling control end of the power supply management unit is connected with an external interface and is used for receiving a digital enabling signal and guaranteeing the temperature stability and reliability of the laser pumping drive circuit.

The application provides a kind of factory multi-device beat coordination control method and system, applied to data processing field, specifically specially applicable to the data processing of supervisory purpose;The application constructs beat sequence by real-time operating parameter of processing equipment, and further establishes multi-device beat coupling topological network, realizes the dynamic analysis of beat correlation and disturbance propagation process between multi-devices, compared with the local speed regulation mode of traditional single-device beat, can identify the propagation path and propagation trend of beat disturbance between multi-devices through beat fluctuation frequency, phase difference and buffer oscillation frequency etc. Beat oscillation characteristics, and accurately locate the source device of beat disturbance in combination with preset phase lead relationship, to avoid periodic amplification of beat fluctuation in production line.

The invention discloses an intelligent prediction system for threshing and separating quality of a soybean combine harvester, and particularly relates to the technical field of intelligent prediction of separating quality. A load attitude coupling space-time description set is constructed, an axial non-uniformity coefficient and a load peak offset are extracted, and a threshing strength axial difference factor is constructed by combining a drum rotating speed; coupling the threshing intensity axial difference factor with the pitch angular velocity to construct an inertial disturbance variable, separating low-frequency rigid body motion and high-frequency threshing characteristic components by using a vibration signal time-frequency decoupling method, and performing distributed prediction on the entrainment loss rate of each axial section by using a time sequence prediction model; on-line self-correction of the model is realized by matching with a dynamic correction factor constructed based on actual measurement deviation and a historical working condition fusion mechanism, and self-adaptive control of threshing energy input is realized by performing closed-loop comparison on the comprehensive quality index and a preset threshold value and adjusting the rotating speed of a roller in a linkage manner.