Bird Flight Trajectory Prediction via Multi-radar Joint Sensing Based on Glitch Filter Algorithm

As fault caused by bird activities on high-voltage towers becomes a major safety problem in transmission line operation, more and more power departments are working with intelligent driving bird devices. Up to now, there are some devices equipped with bird detection technology, but the effect of bird driving is not satisfactory. Therefore, in this paper, a weight computing algorithm based on glitch filter is proposed to determine the sensing degree of Doppler radar, and the results of this algorithm are used to predict bird flight trajectory. First, a multi-radar detection device is designed with the detection space being split. Second, the weight computing algorithm is introduced, and an optimization algorithm is proposed. Then by using a small UAV to perform experiments, a large amount of simulation data is collected. Finally, logistic regression algorithm is applied for verification and analysis, and the obtained experimental results show that weighted computing algorithm and optimization algorithm proposed in this paper have high reliabilitys.