Search results

This paper aims to realize a fully distributed multi-UAV collision detection and avoidance based on deep reinforcement learning (DRL). To deal with the problem of low sample efficiency in DRL and speed up the training. To improve the applicability and reliability of the DRL-based approach in multi-UAV control problems.

In this paper, a fully distributed collision detection and avoidance approach for multi-UAV based on DRL is proposed. A method that integrates human experience into policy training via a human experience-based adviser is proposed. The authors propose a hybrid control method which combines the learning-based policy with traditional model-based control. Extensive experiments including simulations, real flights and comparative experiments are conducted to evaluate the performance of the approach.

A fully distributed multi-UAV collision detection and avoidance method based on DRL is realized. The reward curve shows that the training process when integrating human experience is significantly accelerated and the mean episode reward is higher than the pure DRL method. The experimental results show that the DRL method with human experience integration has a significant improvement than the pure DRL method for multi-UAV collision detection and avoidance. Moreover, the safer flight brought by the hybrid control method has also been validated.

The fully distributed architecture is suitable for large-scale unmanned aerial vehicle (UAV) swarms and real applications. The DRL method with human experience integration has significantly accelerated the training compared to the pure DRL method. The proposed hybrid control strategy makes up for the shortcomings of two-dimensional light detection and ranging and other puzzles in applications.

Control of manipulator collision avoidance by using computer‐aided potential field method is studied. A repulsive force is artificially created using the distances between the robot links and obstacles, which are generated by a distance computation algorithm. Typical approaches for definition of link coordinate frames and kinematics computations are adopted. Control gains in the potential field control model are selected with respect to the ranges of the angular velocity and angular acceleration of an industrial robot. Results on collision detection and collision avoidance control are presented. Using a high‐speed personal computer, real‐time manipulator control was achieved. From the results, changing repulsive force gain can change the safety distance. This makes the safety zone adjustable and provides greater intelligence for robotic tasks under the ever‐changing environment.

The purpose of this paper is to describe simulation research carried out for the needs of multi-sensor anti-collision system for light aircraft and unmanned aerial vehicles.

This paper presents an analysis related to the practical possibilities of detecting intruders in the air space with the use of optoelectronic sensors. The theoretical part determines the influence of the angle of view, distance from the intruder and the resolution of the camera on the ability to detect objects with different linear dimensions. It has been assumed that the detection will be effective for objects represented by at least four pixels (arranged in a line) on the sensor matrix. In the main part devoted to simulation studies, the theoretical data was compared to the obtained intruders’ images. The verified simulation environment was then applied to the image processing algorithms developed for the anti-collision system.

A simulation environment was obtained enabling reliable tests of the anti-collision system using optoelectronic sensors.

The integration of unmanned aircraft operations in civil airspace is a serious problem on a global scale. Equipping aircraft with autonomous anti-collision systems can help solve key problems. The use of simulation techniques in the process of testing anti-collision systems allows the implementation of test scenarios that may be burdened with too much risk in real flights.

This paper aims for possible improvement of safety in light-sport aviation.

This paper conducts verification of classic flight simulator software suitability for carrying out anti-collision systems tests and development of a flight simulator platform dedicated to such tests.

The purpose of this paper is to propose a novel curvilinear path estimation model employing multivariate adaptive regression splines (MARS) for mid vehicle collision avoidance. The two-phase path estimation scheme initially uses the offset (position) value of the front and the mid (host) vehicle to build the crisp model. The resulting crisp model is MARS regressed to deliver a closely aligned actual model in the second phase. This arrangement significantly narrows the gap between the estimated and the true path analyzed using the mean square error (MSE) for different offsets on Next Generation Simulation Interstate 80 (NGSIM I-80) data set. The presented model also covers parallel parking by encompassing the reverse motion of the host vehicle in the path estimation, thereby, making it amicable for real-road scenarios.

The two-phase path estimation scheme initially uses the offset (position) value of the front and the mid (host) vehicle to build the crisp model. The resulting crisp model is MARS regressed to deliver a closely aligned actual model in the second phase.

This arrangement significantly narrows the gap between the estimated and the true path studied using MSE for different offsets on real (Next Generation Simulation-NGSIM) data. The presented model also covers parallel parking by encompassing the reverse motion of the host vehicle in the path estimation. Thereby, making it amicable for real-road scenarios.

This paper builds a mathematical model that considers the offset and host (mid) vehicles for appropriate path fitting.

Collision avoidance is considered as a crucial issue in mobile robotic navigation to guarantee the safety of robots as well as working surroundings, especially for humans. Therefore, the position and velocity of obstacles appearing in the working space of the self-driving mobile robot should be observed to help the robot predict the collision and choose traversable directions. This paper aims to propose a new approach for obstacle tracking, dubbed MoDeT.

First, all long lines, such as walls, are extracted from the 2D-laser scan and considered as static obstacles (or mapped obstacles). Second, a density-based procedure is implemented to cluster nonwall obstacles. These clusters are then geometrically fitted as ellipses. Finally, the combination of Kalman filter and global nearest-neighbor (GNN) method is used to track obstacles’ position and velocity.

The proposed method (MoDeT) is experimentally verified by using an autonomous mobile robot (AMR) named AMR SR300. The MoDeT is found to provide better performance in comparison with previous methods for self-driving mobile robots.

The robot can only see a part of the object, depending on the light detection and ranging scan view. As a consequence, geometrical features of the obstacle are sometimes changed, especially when the robot is moving fast.

This proposed method is to serve the navigation and path planning for the AMR.

(a) Proposing an extended weighted line extractor, (b) proposing a density-based obstacle detection and (c) implementing a combination of methods [in (a) and (b) constant acceleration Kalman and GNN] to obtain obstacles’ properties.

This paper presents a new technique for collision detection between fabric and fabric or fabric and body, applied on our physical‐based fabric drape model. The technique can produce a realistic 3D virtual fashion show based on fabric mechanical properties and has the ability to handle the collision of clothing with an animated synthetic human. The collision technique appeared efficient and reliable when dealing with complex cases of fabric deformation. A full implementation of the drape model, collision detection with an animated human model is also described and discussed.

The purpose of this paper is to provide an overview of unmanned aerial vehicle (UAV) developments, types, the major functional components of UAV, challenges, and trends of UAVs, and among the various challenges, the authors are concentrating more on obstacle sensing methods. This also highlights the scope of on-board vision-based obstacle sensing for miniature UAVs.

The paper initially discusses the basic functional elements of UAV, then considers the different challenges faced by UAV designers. The authors have narrowed down the study on obstacle detection and sensing methods for autonomous operation.

Among the various existing obstacle sensing techniques, on-board vision-based obstacle detection has better scope in the future requirements of miniature UAVs to make it completely autonomous.

The paper gives original review points by doing a thorough literature survey on various obstacle sensing techniques used for UAVs.

Virtual maintenance simulation is of great importance to help designers find and avoid design problems. During its simulation phase, besides the high precision requirement, collision detection must be suitable for all irregular objects in a virtual maintenance environment. Therefore, in this paper, a collision detection approach is proposed based on encapsulation for irregular objects in the virtual maintenance environment.

First, virtual maintenance simulation characteristics and several commonly used bounding boxes methods are analyzed, which motivates the application of encapsulation theory. Based on these, three different encapsulation methods are oriented to the needs of simulation, including encapsulation of rigid maintenance objects, flexible maintenance objects and maintenance personnel. In addition, to detecting collisions accurately, this paper divides the detection process into two stages. That is, in the first stage, a rough detection is carried out and then a tiny slice space is constructed to generate corresponding capsule groups, which will be redetected in the secondary stage. At last, several case studies are applied to illustrate the performance of the methodology.

The automatic construction algorithm for bounding boxes can be adapted to all forms of objects. The number of detection primitives are greatly reduced. It introduces the reachable space of the human body in maintainability as the collision search area.

The advantages of virtual maintenance simulation could also be advantageous in the industry with further studies. The paper believes this study is of particular interest to the readers of your journal.

Outlines the methods and costs of converting existing robotic systems to new applications and gives research results of software and hardware conversion strategies which speed up re‐configuration. Discusses the trade‐offs between buying new solutions or upgrading old ones, describes the uses and extra costs of multi‐arm systems and attempts to generalize the applications to which they can be put. Concludes that the application of co‐operating, multi‐arm systems can provide high system reusability, although this is at the cost of greater planning needs, some of which have been met. Continued enhancement of processing power, sensing and operator interfaces are permitting multi‐arm systems to be used and reused in versatile, high dexterity industrial applications.

This study aims to address the challenge of automatic guided vehicle (AGV) scheduling for parcel storage and retrieval in an intelligent warehouse.

This study presents a scheduling solution that aims to minimize the maximum completion time for the AGV scheduling problem in an intelligent warehouse. First, a mixed-integer linear programming model is established, followed by the proposal of a novel genetic algorithm to solve the scheduling problem of multiple AGVs. The improved algorithm includes operations such as the initial population optimization of picking up goods based on the principle of the nearest distance, adaptive crossover operation evolving with iteration, mutation operation of equivalent exchange and an algorithm restart strategy to expand search ability and avoid falling into a local optimal solution. Moreover, the routing rules of AGV are described.

By conducting a series of comparative experiments based on the actual package flow situation of an intelligent warehouse, the results demonstrate that the proposed genetic algorithm in this study outperforms existing algorithms, and can produce better solutions for the AGV scheduling problem.

This paper optimizes the different iterative steps of the genetic algorithm and designs an improved genetic algorithm, which is more suitable for solving the AGV scheduling problem in the warehouse. In addition, a path collision avoidance strategy that matches the algorithm is proposed, making this research more applicable to real-world scheduling environments.