MNNMs Integrated Control for UAV Autonomous Tracking Randomly Moving Target Based on Learning Method.

Mingjun Li,Kelin Lu,Zhihao Cai,Jiang Zhao,Yingxun Wang,Yibo Wang

doi:10.3390/s21217307

Mingjun Li, Kelin Lu + Show 4 more

Open Access

https://doi.org/10.3390/s21217307

Copy DOI

Abstract

In this paper, we investigate the problem of unmanned aerial vehicles (UAVs) autonomous tracking moving target with only an airborne camera sensor. We proposed a novel integrated controller framework for this problem based on multi-neural-network modules (MNNMs). In this framework, two neural networks are designed for target perception and guidance control, respectively. The deep learning method and reinforcement learning method are applied to train the integrated controller. The training result demonstrates that the integrated controller can be trained more quickly and efficiently than the end-to-end controller trained by the deep reinforcement learning method. The flight tests with the integrated controller are implemented in simulated and realistic environments, the results show that the integrated controller trained in simulation can easily be transferred to the realistic environment and achieve the UAV tracking randomly moving target, which has a faster motion velocity. The integrated controller based on the MNNMs structure has a better performance on an autonomous tracking target than the control mode that combines with a perception network and a proportional integral derivative controller.

Highlights

IntroductionThe rapid development of unmanned aerial vehicles (UAVs) technology has led to the widespread application of UAVs in a variety of fields and missions, such as natural disaster rescue [1], environment exploration [2], and military reconnaissance [3]
In recent years, the rapid development of unmanned aerial vehicles (UAVs) technology has led to the widespread application of UAVs in a variety of fields and missions, such as natural disaster rescue [1], environment exploration [2], and military reconnaissance [3].In many application scenarios, the UAVs are required to have the ability to track moving targets [4]
We propose an integrated neural network framework based on multineural-network modules (MNNMs) for a UAV autonomous tracking moving target

Summary

Introduction

The rapid development of unmanned aerial vehicles (UAVs) technology has led to the widespread application of UAVs in a variety of fields and missions, such as natural disaster rescue [1], environment exploration [2], and military reconnaissance [3]. The UAVs are required to have the ability to track moving targets [4]. As the mission environment for UAVs has become increasingly complex and challenging, the UAV technology will be further developed toward autonomous control with limited payload and information [5]. We consider the problem of UAV autonomous tracking moving targets with a single camera sensor. The PID controller cannot cope with various tracking situations, and for each case, it requires tedious gain tuning to obtain satisfactory results

Methods

Results

Conclusion