Algorithm For Unmanned Aerial Vehicle Research Articles

A novel reinforcement learning based grey wolf optimizer algorithm for unmanned aerial vehicles (UAVs) path planning

Unmanned aerial vehicles (UAVs) have been used in wide range of areas, and a high-quality path planning method is needed for UAVs to satisfy their applications. However, many algorithms reported in the literature may not feasible or efficient, especially in the face of three-dimensional complex flight environment. In this paper, a novel reinforcement learning based grey wolf optimizer algorithm called RLGWO has been presented for solving this problem. In the proposed algorithm, the reinforcement learning is inserted that the individual is controlled to switch operations adaptively according to the accumulated performance. Considering that the proposed algorithm is designed to serve for UAVs path planning, four operations have been introduced for each individual: exploration, exploitation, geometric adjustment, and optimal adjustment. In addition, the cubic B-spline curve is used to smooth the generated flight route and make the planning path be suitable for the UAVs. The simulation experimental results show that the RLGWO algorithm can acquire a feasible and effective route successfully in complicated environment.

Co-operative Collision Avoidance for Unmanned Aerial Vehicles using both Centralised and Decoupled Approaches

Abstract This paper presents the design of co-operative collision avoidance algorithms for Unmanned Aerial Vehicles (UAVs) using vertical avoidance manoeuvres. The co-operative collision avoidance problem is formulated as an optimal control problem and solved using an A⁎ search algorithm. Two different approaches are developed and compared: a centralised approach where the collision avoidance trajectories for all UAVs are planned simultaneously, and a decoupled approach where the individual collision avoidance trajectories for each UAV is planned sequentially, with the planning sequence determined by a UAV priority order. The UAVs co-operate by sharing their state and intent information with one another and with a central node, if present. The co-operative collision avoidance algorithms are verified and evaluated using illustrative simulations. These simulations support the expected behaviour of the algorithms. The centralised approach finds the most optimal solution to the problem while the solution found by the decoupled approach depends on the priority allocation of the UAVs. The decoupled approach can find either the most optimal or a sub-optimal solution to the problem, with the priority allocation occasionally resulting in the decoupled approach being unable to find a solution. This suggests that the centralised approach will, on average, find solutions more often and find more optimal solutions than the decoupled approach.

Reactive Collision Avoidance Algorithm for UAV Using Bounding Tube Against Multiple Moving Obstacles

A reactive collision avoidance algorithm is proposed to enable safe autonomous flight in the presence of multiple dynamic obstacles. A position-controlled hexacopter equipped with a visual sensor obtaining obstacle information is considered as an Unmanned Aerial Vehicle (UAV) platform. The proposed method centers on the concept of bounding tube which intrinsically extends the static bounding box to incorporate forthcoming movement of the obstacles into the collision avoidance framework. The processing pipeline consists of separate components for each of the sequential tasks in obstacle sensing and tracking. Computation of a spherical bounding box for each obstacle is followed by discrete-time Kalman filtering for prediction of obstacle trajectory to detect potential collision. If the current course of UAV turns out highly likely to end in collision with any of the obstacles, the vehicle steers to an aiming point chosen from among the bundle of candidates produced by constructing a bounding tube that takes account of predicted obstacle motion. The bounding-tube-based aiming point generation extends seamlessly to the case with multiple moving obstacles through running in series with multi-obstacle track management that combines hierarchical clustering of sensory data points for obstacle identification and a simple geometric method for data association. Numerical simulations are conducted to verify the performance of the proposed collision avoidance algorithm.

Combination of Wide Baseline Image Matching and Tracking for Autonomous UAV Approaches to a Window

In this paper, the problem of an autonomous unmanned aerial vehicle (UAV) approach to a target which is selected in a reference image is addressed. A robust matching algorithm is proposed to reliably project the selected point in the reference image into the live images of a quadrotor helicopter. Projective transformations are applied to the reference image to extract additional keypoints and to gain invariance to out-of-plane perspective transformations. Since the matching algorithm does not exploit the beneficial characteristics of image sequences and because its processing time is not short enough for high frame rates, a tracking algorithm is introduced. High detection rates even for image sequences with large viewpoint changes are achieved. Therefore, the presented algorithm can be used as input to a guidance algorithm for UAVs.

Path Planning Algorithms for Unmanned Aerial Vehicles

Path Planning Algorithms for Unmanned Aerial Vehicles

Damping oscillation of suspended payload by up and down motion of the pivot base - time delay algorithms for UAV applications

A case study problem of damping oscillations of a pendulum by moving a pivot in vertical direction is presented. The motivation for solving this task comes from a broader research on designing advanced control algorithms for Unmanned Aerial Vehicles (UAVs) carrying a suspended payload. First, a nonlinear model for vertical motion of a quadcopter carrying a payload is derived and simplified to a nonlinear pendulum-like model. The main result is in deriving a nonlinear time-delay control law allowing to predefine a damping factor of the oscillations. Note that the problem at hand cannot be solved by a linear controller, as the oscillatory mode is not controllable once the model is linearised. The control law is then extended from pendulum to the quadcopter model considering an outer loop to control the horizontal motion. The theoretical results are then validated by simulations.

基于改进ORB和PROSAC的无人机航拍图像拼接算法

基于改进ORB和PROSAC的无人机航拍图像拼接算法

Collaborative Consistency Control Algorithm Research of Unmanned Aerial Vehicle Formation

In this paper, the cooperative consistency control algorithm of unmanned aerial vehicle is studied. The formation of unmanned aerial vehicles has the characteristics of controlling uncertainty. This makes the controller have certain uncertainty which may cause the multi-agent system to be difficult to agree. In view of this feature, the intelligent algorithm for collaborative consistency control of unmanned aerial vehicle formation is designed with the properties of the Laplace matrix of the multi-agent network which is composed of a series of intelligent bodies with autonomous capability and the interaction between them. The simulation results verify the feasibility of the intelligent algorithm.

Low-Complexity Path Planning Algorithm for Unmanned Aerial Vehicles in Complicated Scenarios

Existing algorithms on path planning with obstacles for unmanned aerial vehicles (UAVs) suffer from high computational complexity and unpredictability when the considered scenario is complicated. In this paper, we propose a novel path-planning algorithm for UAVs, which relies on continuously updating virtual regional field and its local gradients. The information of target regions and obstacles is incorporated in a virtual regional field. The algorithm circumvents the large number of variables to be optimized, and does not rely on any black boxes with unpredictable outputs. Real data show that the proposed algorithm can design a path with high coverage rate of the target region in a certain time duration, and guides the UAV to bypass the obstacles. The approach based on the regional field provides an option for low-cost hardwares, and reveals insights into the problem of path planning.

Collision avoidance algorithms for unmanned aerial vehicles using computer vision

We present a computer-vision-based approach to enable unmanned aerial vehicles (UAVs) to avoid collisions. To detect a moving obstacle, a machine learning algorithm, cascade classification, is used. Next, to track the obstacle, the camshift algorithm is implemented. These algorithms determine the coordinates of center of an object moving towards the UAV. Finally, to determine the distance of the object to the UAV, a stereo camera is used. Once an object is successfully tracked, the UAV will execute avoidance maneuvers if the path of the moving object conflicts with the UAV's path.

Distributed Control for Groups of Unmanned Aerial Vehicles Performing Surveillance Missions and Providing Relay Communication Network Services

This article presents the development of an autonomous and distributed movement coordination algorithm for Unmanned Aerial Vehicles (UAVs) swarms used in communication relay networks and in exploratory area surveillance missions. This work studies the performance of a hybrid algorithm combining pheromone maps, market auction paradigms and proactive link maintenance mechanisms to create a self-organizing flying network capable of providing network support for the UAV nodes already engaged in exploration and targeting tasks in the surveillance missions. In order to validate the proposal, simulations were performed assessing the desired performance aspects related to the target allocation and network connectivity. The acquired results provide evidence that the proposed solution is able to maintain the balance between the performance goals.

Novel Levenberg–Marquardt based learning algorithm for unmanned aerial vehicles

In this paper, Levenberg–Marquardt inspired sliding mode control theory based adaptation laws are proposed to train an intelligent fuzzy neural network controller for a quadrotor aircraft. The proposed controller is used to control and stabilize a quadrotor unmanned aerial vehicle in the presence of periodic wind gust. A proportional-derivative controller is firstly introduced based on which fuzzy neural network is able to learn the quadrotor’s control model on-line. The proposed design allows handling uncertainties and lack of modelling at a computationally inexpensive cost. The parameter update rules of the learning algorithms are derived based on a Levenberg–Marquardt inspired approach, and the proof of the stability of two proposed control laws are verified by using the Lyapunov stability theory. In order to evaluate the performance of the proposed controllers extensive simulations and real-time experiments are conducted. The 3D trajectory tracking problem for a quadrotor is considered in the presence of time-varying wind conditions.

Use of Co-operative UAVs to Support/Augment UGV Situational Awareness and/or Inter-Vehicle Communications

This paper presents the development of a path-planning algorithm for Unmanned Aerial Vehicles (UAVs) in order to increase the situational awareness for platooning vehicles. The scenario considers a team of cooperative UAVs, initially docked on moving Unmanned Ground Vehicles (UGVs). In particular, the goal consists in finding the best routing plan for the UAVs in order to visit some designated targets in a wide search area to augment the UGVs’ situational awareness. Taking into account the maximal endurance constraint of the UAVs, this problem becomes equivalent to a time-constrained multiple depot vehicle routing problem with moving depots. To tackle this variant of the well-known vehicle routing problem, a methodology based on a TABU search meta-heuristic is implemented. While respecting the endurance constraint, this methodology tempts to optimize multiple objectives: the number of UAVs required, the total length of the routing and the balance of the lengths of the different routes within the routing plan. The proposed approach based on a meta-heuristic gives relevant results in a relatively short period of time.

Trajectory generation using reinforcement learning for autonomous helicopter with adaptive dynamic movement primitive

The present paper introduces a smart trajectories generation algorithm for unmanned aerial vehicles under various environments. Dynamic movement primitive is extended by adding jerk to mock the kinematics, particularly for unmanned aerial vehicles. Combining the improved dynamic movement primitive with policy learning by weighted exploration with the returns, we propose the new algorithm producing optimal trajectories under different scenarios. Furthermore, numerical simulations under several scenarios are performed, demonstrating the ability of the proposed algorithm.

Trajectory Planning for Communication Relay Unmanned Aerial Vehicles in Urban Dynamic Environments

This paper proposes an optimal positioning and trajectory planning algorithm for unmanned aerial vehicles (UAVs) to improve a communication quality of a team of ground mobile nodes (vehicles) in a complex urban environment. In particular, a nonlinear model predictive control (NMPC)-based approach is proposed to find an efficient trajectory for UAVs with a discrete genetic algorithm while considering the dynamic constraints of fixed-wing UAVs. The advantages of using the proposed NMPC approach and the communication performance metrics are investigated through a number of scenarios with different horizon steps in the NMPC framework, the number of UAVs used, heading rates and speeds.

Adaptive sensitivity decision based path planning algorithm for unmanned aerial vehicle with improved particle swarm optimization

Automatic path planning is an essential aspect of unmanned aerial vehicle (UAV) autonomy. This paper presents a three dimensional path planning algorithm based on adaptive sensitivity decision operator combined with particle swarm optimization (PSO) technique. In the proposed method, an adaptive sensitivity decision area is constructed to overcome the defects of local optimal and slow convergence. By using this specified area, the potential particle locations with high probabilities are determined and other candidates are deleted to improve computational capacity. Then the searching space of particles is constrained in a limited boundary to avoid premature state. In addition, the searching accuracy is enhanced by the relative particle directivity from current location. The objective function is redesigned by taking into account the distance to destination and UAV self-constraints. To evaluate the path length, the paired-sample T-Test is performed and the straight line rate (SLR) index is introduced. In the two scenarios applied in this paper, our proposed method is 35.4%, 21.6% and 49.5% better compared with other three tested optimization algorithms in the path cost on average. Correspondingly it is 9.6%, 12.8%, and 25.3% better in SLR, which is capable of generating higher quality paths efficiently for UAVs.

A Hybrid Filtering Approach of Digital Video Stabilization for UAV Using Kalman and Low Pass Filter

In this paper a new video stabilization algorithm for unmanned aerial vehicles (UAV) has been presented which is used to stabilize the video being transmitted from UAV to the ground station. First, the corner points are extracted using Good Features to Track corner detection algorithm and the extracted points are used to compute the optical flow between two consecutive frames. Next, the points detected from optical flow are used to estimate the motion parameters using an affine transform model. Subsequently, a hybrid filter consisting of Kalman and low pass filter is used to smooth the estimated motion parameters and the frames are warped using the smoothed parameters to obtain a stabilized video sequence. The experimental results show that the algorithm can remove the unwanted vibration more effectively than the one that only uses either a Kalman Filter or a low pass filter.

Swarming Algorithm for Unmanned Aerial Vehicle (UAV) Quadrotors – Swarm Behavior for Aggregation, Foraging, Formation, and Tracking –

This paper presents the fusion of swarm behavior in multi robotic system specifically the quadrotors unmanned aerial vehicle (QUAV) operations. This study directed on using robot swarms because of its key feature of decentralized processing amongst its member. This characteristic leads to advantages of robot operations because an individual robot failure will not affect the group performance. The algorithm emulating the animal or insect swarm behaviors is presented in this paper and implemented into an artificial robotic agent (QUAV) in computer simulations. The simulation results concluded that for increasing number of QUAV the aggregation accuracy increases with an accuracy of 90.62%. The experiment for foraging revealed that the number of QUAV does not affect the accuracy of the swarm instead the iterations needed are greatly improved with an average of 160.53 iterations from 50 to 500 QUAV. For swarm tracking, the average accuracy is 89.23%. The accuracy of the swarm formation is 84.65%. These results clearly defined that the swarm system is accurate enough to perform the tasks and robust in any QUAV number.

Flight Conflict Detection Algorithm for UAV and MAV Under the Whole Airspace

The distributed market mechanism based on agent’s BDI model (Believe, Desire, Intention, BDI) is applied to flight conflict detection about UAV (Unmanned Aerial Vehicle, UAV) and MAV (Manned Aerial Vehicle, MAV) under the whole airspace. The state between announce agent and bid agent was changed by the BDI action driving rules. In order to reduce the range of bid agents, the threat operator is put forward to judge flight conflict. Furthermore, a new computational method of bidding value is set up by the vector of relative velocity. The simulation results show that distributed market mechanism can allocate the most threatening UAV to corresponding MAV and the time efficiency analysis indicates that it has better time efficiency than using threat matrix.

UAV Collision Detection Algorithm Design Based on Circle Overlapping Test

Each UAV require a robust collision detection in its operating system to ensure it can aware the potential of collision in near future. Collision detection algorithm is one of the important parts that need to be concerned in the CAS. This paper proposes a design of collision avoidance algorithm for unmanned aerial vehicle based on circle overlapping test. Assuming two moving UAV are flying in the same altitude and in straight path within a two dimensional plane as conflict problem. Both UAV are cooperative aircrafts which communication of position, heading angle, waypoint, and velocity are allowed. The proposed algorithm will determine the collision potential between the surrounding intruders according to the conflict information.