Owing to its powerful adaptability and robustness to the weak light, the infrared camera equipped on the unmanned aerial vehicle is more and more applied for aerial photography. Nowadays, how to make full use of the aerial infrared image sequences for the moving vehicle detection has attracted widespread attention. However, due to the low-resolution, low contrast, and few texture features of the infrared image, it is extremely difficult to detect the moving vehicles. In our work, an effective and efficient moving vehicle detection approach in the aerial infrared image sequences is proposed via fast image registration and You Only Look Once Version 3 (YOLOv3) network. First, to compensate for the motion of the aerial infrared camera, a fast infrared image registration method is put forward. To improve the accuracy and efficiency of image registration, we construct a multi-screening based mechanism for screening out the incorrect and redundant feature points. For feature description, the low-level and high-level descriptors are combined to further improve the registration accuracy. Then, the 2-frame difference and image masking are introduced to acquire the frame mask images, where only the region-of-interest is reserved, and the remaining regions are masked. Next, we construct a new structure of an improved YOLOv3 network with only 23 layers. Due to the insufficiency of the infrared vehicle samples, transfer learning is introduced to train the improved YOLOv3 network. Finally, the proposed approach is evaluated on the Defence Advanced Research Projects Agency (DARPA) Video verification of Identify (VIVID) and Northwest Polytechnical University (NPU) data sets. Experiments and comprehensive analyses show that the proposed approach can achieve satisfactory and competitive moving vehicle detection results.
1 July 2017 in Frontiers of Information Technology & Electronic Engineering
We propose a formation control strategy for multiple unmanned aerial vehicles (multi-UAV) based on second-order consensus, by introducing position and velocity coordination variables through neighbor-to-neighbor interaction to generate steering commands. A cooperative guidance algorithm and a cooperative control algorithm are proposed together to maintain a specified geometric configuration, managing the position and attitude respectively. With the whole system composed of the six-degree-of-freedom UAV model, the cooperative guidance algorithm, and the cooperative control algorithm, the formation control strategy is a closed-loop one and with full states. The cooperative guidance law is a second-order consensus algorithm, providing the desired acceleration, pitch rate, and heading rate. Longitudinal and lateral motions are jointly considered, and the cooperative control law is designed by deducing state equations. Closed-loop stability of the formation is analyzed, and a necessary and sufficient condition is provided. Measurement errors in position data are suppressed by synchronization technology to improve the control precision. In the simulation, three-dimensional formation flight demonstrates the feasibility and effectiveness of the formation control strategy.
This paper develops a flexible collision avoidance strategy for the formation of multiple unmanned aerial vehicles (multi-UAV). Firstly, To make full use of the information flow from the communication of the formation, an improved artificial potential field (IAPF) function is constructed by adding the communication topology and communication weights. Based on the IAPF functions, both collision avoidance within the formation and obstacle avoidance outside the formation are well organized to ensure flight safety. Secondly, to fuse collision avoidance and formation keeping, we design the null space behavioral (NSB) approach to combine them into a single motion command, where the formation keeping algorithm is designed as a second-order consensus algorithm. In this way, the flexible collision avoidance strategy is proposed based on IAPF and consensus, handling the relationship between collision avoidance and formation keeping flexibly. Finally, three-dimensional multi-UAV flight simulation validates the effectiveness of the proposed strategy that the formation can both implement collision avoidance quickly and keep a good geometric formation configuration simultaneously.
Vehicle detection in aerial images has attracted great attention as an approach to providing the necessary information for transportation road network planning and traffic management. However, because of the low resolution, complex scene, occlusion, shadows, and high requirement for detection efficiency, implementing vehicle detection in aerial images is challenging. Therefore, we propose an efficient and scene-adaptive algorithm for vehicle detection in aerial images using an improved YOLOv3 framework, and it is applied to not only aerial still images but also videos composed of consecutive frame images. First, rather than directly using the traditional YOLOv3 network, we construct a new structure with fewer layers to improve the detection efficiency. Then, since complex scenes in aerial images can cause the partial occlusion of vehicles, we construct a context-aware-based feature map fusion to make full use of the information in the adjacent frames and accurately detect partially occluded vehicles. The traditional YOLOv3 network adopts a horizontal bounding box, which can attain the expected detection effects only for vehicles with small length–width ratio. Moreover, vehicles that are close to each other are liable to cause lower accuracy and a higher detection error rate. Hence, we design a sloping bounding box attached to the angle of the target vehicles. This modification is conducive to predicting not only the position but also the angle. Finally, two data sets were used to perform extensive experiments and comparisons. The results show that the proposed algorithm generates the desired and excellent performance.
1 June 2017 in Journal of Central South University
A distributed coordinated consensus problem for multiple networked Euler-Lagrange systems is studied. The communication between agents is subject to time delays, unknown parameters and nonlinear inputs, but only with their states available for measurement. When the communication topology of the system is connected, an adaptive control algorithm with self-delays and uncertainties is suggested to guarantee global full-state synchronization that the difference between the agent's positions and velocities asymptotically converges to zero. Moreover, the distributed sliding-mode law is given for chaotic systems with nonlinear inputs to compensate for the effects of nonlinearity. Finally, simulation results show the effectiveness of the proposed control algorithm.