Multi-Robot Collaborative Hunting in Cluttered Environments With Obstacle-Avoiding Voronoi Cells

This work proposes an online collaborative hunting strategy for multi-robot systems based on obstacle-avoiding Voronoi cells in a complex dynamic environment. This involves firstly designing the construction method using a support vector machine(SVM) based on the definition of buffered Voronoi cells(BVCs). Based on the safe collision-free region of the robots, the boundary weights between the robots and the obstacles are dynamically updated such that the robots are tangent to the buffered Voronoi safety areas without intersecting with the obstacles. Then, the robots are controlled to move within their own buffered Voronoi safety area to achieve collision-avoidance with other robots and obstacles. The next step involves proposing a hunting method that optimizes collaboration between the pursuers and evaders. Some hunting points are generated and distributed evenly around a circle. Next, the pursuers are assigned to match the optimal points based on the Hungarian algorithm.Then, a hunting controller is designed to improve the containment capability and minimize containment time based on collision risk. Finally, simulation results have demonstrated that the proposed cooperative hunting method is more competitive in terms of time and travel distance.

Hierarchical adaptive stereo matching algorithm for obstacle detection with dynamic programming

An adaptive weighted stereo matching algorithm with multilevel and bidirectional dynamic programming based on ground control points （GCPs） is presented. To decrease time complexity without losing matching precision, using a multilevel search scheme, the coarse matching is processed in typical disparity space image, while the fine matching is processed in disparity-offset space image. In the upper level, GCPs are obtained by enhanced volumetric iterative algorithm enforcing the mutual constraint and the threshold constraint. Under the supervision of the highly reliable GCPs, bidirectional dynamic programming framework is employed to solve the inconsistency in the optimization path. In the lower level, to reduce running time, disparity-offset space is proposed to efficiently achieve the dense disparity image. In addition, an adaptive dual support-weight strategy is presented to aggregate matching cost, which considers photometric and geometric information. Further, post-processing algorithm can ameliorate disparity results in areas with depth discontinuities and related by occlusions using dual threshold algorithm, where missing stereo information is substituted from surrounding regions. To demonstrate the effectiveness of the algorithm, we present the two groups of experimental results for four widely used standard stereo data sets, including discussion on performance and comparison with other methods, which show that the algorithm has not only a fast speed, but also significantly improves the efficiency of holistic optimization.

Dynamic A^*path finding algorithm and 3D lidar based obstacle avoidance strategy for autonomous vehicles

This paper presents a novel dynamic A^*path finding algorithm and 3D lidar based local obstacle avoidance strategy for an autonomous vehicle.3D point cloud data is collected and analyzed in real time.Local obstacles are detected online and a 2D local obstacle grid map is constructed at 10 Hz/s.The A^*path finding algorithm is employed to generate a local path in this local obstacle grid map by considering both the target position and obstacles.The vehicle avoids obstacles under the guidance of the generated local path.Experiment results have shown the effectiveness of the obstacle avoidance navigation algorithm proposed.

Swarm intelligence based dynamic obstacle avoidance for mobile robots under unknown environment using WSN

To solve dynamic obstacle avoidance problems, a novel algorithm was put forward with the advantages of wireless sensor network (WSN). In view of moving velocity and direction of both the obstacles and robots, a mathematic model was built based on the exposure model, exposure direction and critical speeds of sensors. Ant colony optimization (ACO) algorithm based on bionic swarm intelligence was used for solution of the multi-objective optimization. Energy consumption and topology of the WSN were also discussed. A practical implementation with real WSN and real mobile robots were carried out. In environment with multiple obstacles, the convergence curve of the shortest path length shows that as iterative generation grows, the length of the shortest path decreases and finally reaches a stable and optimal value. Comparisons show that using sensor information fusion can greatly improve the accuracy in comparison with single sensor. The successful path of robots without collision validates the efficiency, stability and accuracy of the proposed algorithm, which is proved to be better than tradition genetic algorithm (GA) for dynamic obstacle avoidance in real time.

Dynamic Frontier-Led Swarming:Multi-Robot Repeated Coverage in Dynamic Environments

A common assumption of coverage path planning research is a static environment.Such environments require only a single visit to each area to achieve coverage.However,some real-world environments are characterised by the presence of unexpected,dynamic obstacles.They require areas to be revisited periodically to maintain an accurate coverage map,as well as reactive obstacle avoidance.This paper proposes a novel swarmbased control algorithm for multi-robot exploration and repeated coverage in environments with unknown,dynamic obstacles.The algorithm combines two elements:frontier-led swarming for driving exploration by a group of robots,and pheromone-based stigmergy for controlling repeated coverage while avoiding obstacles.We tested the performance of our approach on heterogeneous and homogeneous groups of mobile robots in different environments.We measure both repeated coverage performance and obstacle avoidance ability.Through a series of comparison experiments,we demonstrate that our proposed strategy has superior performance to recently presented multi-robot repeated coverage methodologies.

Path Planning of UAV by Combing Improved Ant Colony System and Dynamic Window Algorithm

A fusion algorithm is proposed to enhance the search speed of an ant colony system(ACS)for the global path planning and overcome the challenges of the local path planning in an unmanned aerial vehicle(UAV).The ACS search efficiency is enhanced by adopting a 16-direction 24-neighborhood search way,a safety grid search way,and an elite hybrid strategy to accelerate global convergence.Quadratic planning is performed using the moving average(MA)method.The fusion algorithm incorporates a dynamic window approach(DWA)to deal with the local path planning,sets a retracement mechanism,and adjusts the evaluation function accordingly.Experimental results in two environments demonstrate that the improved ant colony system(IACS)achieves superior planning efficiency.Additionally,the optimized dynamic window approach(ODWA)demonstrates its ability to handle multiple dynamic situations.Overall,the fusion optimization algorithm can accomplish the mixed path planning effectively.

A dynamic fusion path planning algorithm for mobile robots incorporating improved IB-RRT∗and deep reinforcement learning

Dynamic path planning is crucial for mobile robots to navigate successfully in unstructured envi-ronments.To achieve globally optimal path and real-time dynamic obstacle avoidance during the movement,a dynamic path planning algorithm incorporating improved IB-RRT∗and deep reinforce-ment learning(DRL)is proposed.Firstly,an improved IB-RRT∗algorithm is proposed for global path planning by combining double elliptic subset sampling and probabilistic central circle target bi-as.Then,to tackle the slow response to dynamic obstacles and inadequate obstacle avoidance of tra-ditional local path planning algorithms,deep reinforcement learning is utilized to predict the move-ment trend of dynamic obstacles,leading to a dynamic fusion path planning.Finally,the simulation and experiment results demonstrate that the proposed improved IB-RRT∗algorithm has higher con-vergence speed and search efficiency compared with traditional Bi-RRT∗,Informed-RRT∗,and IB-RRT∗algorithms.Furthermore,the proposed fusion algorithm can effectively perform real-time obsta-cle avoidance and navigation tasks for mobile robots in unstructured environments.

On-line real-time path planning of mobile robots in dynamic uncertain environment

A new path planning method for mobile robots in globally unknown environment with moving obstacles is pre- sented. With an autoregressive (AR) model to predict the future positions of moving obstacles, and the predicted position taken as the next position of moving obstacles, a motion path in dynamic uncertain environment is planned by means of an on-line real-time path planning technique based on polar coordinates in which the desirable direction angle is taken into consideration as an optimization index. The effectiveness, feasibility, high stability, perfect performance of obstacle avoidance, real-time and optimization capability are demonstrated by simulation examples.

Neural network-based source tracking of chemical leaks with obstacles

The leakage of hazardous gases poses a significant threat to public security and causes environmental damage.The effective and accurate source term estimation(STE)is necessary when a leakage accident occurs.However,most research generally assumes that no obstacles exist near the leak source,which is inappropriate in practical applications.To solve this problem,we propose two different frameworks to emphasize STE with obstacles based on artificial neural network(ANN)and convolutional neural network(CNN).Firstly,we build a CFD model to simulate the gas diffusion in obstacle scenarios and construct a benchmark dataset.Secondly,we define the structure of ANN by searching,then predict the concentration distribution of gas using the searched model,and optimize source term parameters by particle swarm optimization(PSO)with well-performed cost functions.Thirdly,we propose a one-step STE method based on CNN,which establishes a link between the concentration distribution and the location of obstacles.Finally,we propose a novel data processing method to process sensor data,which maps the concentration information into feature channels.The comprehensive experiments illustrate the performance and efficiency of the proposed methods.

Path Planning of Quadrotors in a Dynamic Environment Using a Multicriteria Multi-Verse Optimizer

Paths planning of Unmanned Aerial Vehicles(UAVs)in a dynamic environment is considered a challenging task in autonomous flight control design.In this work,an efficient method based on a Multi-Objective MultiVerse Optimization(MOMVO)algorithm is proposed and successfully applied to solve the path planning problem of quadrotors with moving obstacles.Such a path planning task is formulated as a multicriteria optimization problem under operational constraints.The proposed MOMVO-based planning approach aims to lead the drone to traverse the shortest path from the starting point and the target without collision with moving obstacles.The vehicle moves to the next position from its current one such that the line joining minimizes the total path length and allows aligning its direction towards the goal.To choose the best compromise solution among all the non-dominated Pareto ones obtained for compromise objectives,the modified Technique for Order Preference by Similarity to Ideal Solution(TOPSIS)is investigated.A set of homologous metaheuristics such as Multiobjective Salp Swarm Algorithm(MSSA),Multi-Objective Grey Wolf Optimizer(MOGWO),Multi-Objective Particle Swarm Optimization(MOPSO),and Non-Dominated Genetic Algorithm II(NSGAII)is used as a basis for the performance comparison.Demonstrative results and statistical analyses show the superiority and effectiveness of the proposed MOMVO-based planning method.The obtained results are satisfactory and encouraging for future practical implementation of the path planning strategy.

采摘机器人的路径规划系统动态性优化研究

为进一步改善采摘机器人的工作性能,提出以动态调控为主导的理念,针对整机的路径规划系统展开优化研究。以当前果园采摘机器人的通用性结构组成为前提,将云平台数据处理与路径规划核心算法有效融合后搭建动态控制模型,分别针对路径规划系统的硬件配置与软件控制进行合理设计,得到可应用于采摘实践且布局完整的路径规划系统。展开动态性优化下的采摘作业试验,结果表明:优化后采摘机器人路径规划系统的整体路径搜索率与路径平滑性得到明显提升,相对提升度分别为10.93%和9.71%,路径偏离率相对降低了50%左右,很好地优化了机器人的避障能力,满足系统稳定性需求,具有较高的实用价值。

融合改进A^(*)算法和动态窗口法的自动驾驶路径规划

针对自动驾驶汽车路径规划全局最优、耗时最优和避障的需求,提出一种改进A^(*)算法和动态窗口法的融合算法。A^(*)算法主要从启发函数、权重系数、搜索邻域和搜索策略4个方面进行改进,动态窗口法主要改进评价函数。利用改进后的A^(*)算法和双向A^(*)算法完成栅格地图上的全局路径规划,去除冗余节点并平滑处理优化全局路径,利用融合动态窗口算法进行局部路径规划,完成避障。与传统的A^(*)算法相比,改进的A^(*)算法和双向A^(*)算法搜索全局路径耗时和节点显著减少,优化的A^(*)算法与动态窗口法的融合算法具有更高的效率、更好的路径规划能力和避障能力。

基于FIA*-APF算法的蟹塘投饵船动态路径规划

为了提高无人投饵船在含障碍物河蟹养殖池塘自主巡航的作业效率和安全性,该研究提出基于改进A*算法与人工势场法相融合(fusion of improved A*and artificial potential field,FIA*-APF)的蟹塘投饵船动态路径规划算法。首先引入动态加权因子优化A*算法评价函数;其次加入转折惩罚函数并删除冗余点,接着利用B样条曲线对全局路径进行平滑处理;最后将改进A*算法得到的全局路径作为改进人工势场法中的引力路径,生成投饵船自主巡航高效路径。根据养殖池塘创建静态和动态2种仿真环境,分别对传统人工势场法(traditional artificial potential field,TAPF)、基于A*和人工势场法的融合算法(the A*and artificial potential field,TA*-APF)和FIA*-APF算法的性能进行20次测试。仿真试验结果表明:2种环境下,FIA*-APF算法的平均规划时间是TAPF算法的17.23%,是TA*-APF算法的51.96%,平均指令节点数量比TAPF算法减少50.64%,比TA*-APF算法减少65.03%,平均路径长度比TA*-APF算法减少2.82%。蟹塘试验结果表明:FIA*-APF算法的规划时间为TAPF算法的38.16%,为TA*-APF的62.42%,路径长度比TAPF算法减少29.13%,比TA*-APF减少10.15%;另外,TAPF和TA*-APF算法规划路径上大于60°的转角分别是FIA*-APF算法的3.28和2.62倍,大于100°的转角分别是FIA*-APF算法的3.73和1.67倍,该研究算法规划的路径更高效平滑。研究结果可为无人投饵船自主导航提供参考。

改进Q-Learning的路径规划算法研究

针对Q-Learning算法学习效率低、收敛速度慢且在动态障碍物的环境下路径规划效果不佳的问题,本文提出一种改进Q-Learning的移动机器人路径规划算法.针对该问题,算法根据概率的突变性引入探索因子来平衡探索和利用以加快学习效率;通过在更新函数中设计深度学习因子以保证算法探索概率;融合遗传算法,避免陷入局部路径最优同时按阶段探索最优迭代步长次数,以减少动态地图探索重复率;最后提取输出的最优路径关键节点采用贝塞尔曲线进行平滑处理,进一步保证路径平滑度和可行性.实验通过栅格法构建地图,对比实验结果表明,改进后的算法效率相较于传统算法在迭代次数和路径上均有较大优化,且能够较好的实现动态地图下的路径规划,进一步验证所提方法的有效性和实用性.

近海复杂环境下UUV动态路径规划方法研究

为解决近海环境下水下无人航行器(unmanned underwater vehicle,UUV)的动态路径规划问题,本文提出一种结合全局和局部动态路径规划的算法。首先,本文提出一种基于自适应目标引导的快速拓展随机树算法,以增加随机树生长的方向性,并通过转向和重选策略减少无效拓展加快算法的收敛速度。接着,获得全局路径之后使用自适应子节点选取策略获取动态窗口法的子目标点,将复杂的全局动态任务规划分解为多个简单的动态路劲规划,从而防止动态窗口法陷入局部极小值。最后,通过UUV出港任务仿真实验验证了算法的有效性和实用性。

基于改进哈里斯鹰优化算法的动态路径规划研究

针对传统栅格地图下的路径规划算法存在多峰值优化、无法实时避障等问题,提出了一种基于改进哈里斯鹰优化算法的动态路径规划方法。首先,提出方形邻格邻近扩散方法初始化哈里斯鹰种群位置,在路径规划问题模型下增加种群多样性;然后,提出一种非线性能量因子优化算法在搜索和开发之间的更新比例,提高全局搜索性能;最后,引入动态窗口法提高机器人实际运行路径的平滑程度,构造结合全局路径的动态窗口评价函数以改善动态窗口法前瞻性不足的问题。实验结果表明,所提方法可以兼顾实时避障和路径最优的需求。

改进粒子群算法的机器人避障偏差控制方法

为针对巡检机器人避障偏差进行良好控制,提升避障效果,提出改进粒子群算法的机器人避障偏差控制方法设计。先分析巡检机器人正向动力学和逆向动力学,获取双走轮坐标系下机器人的运动情况,建立机器人运动学方程。然后以此为基础,在双走轮坐标系中,通过改进粒子群算法确定巡检机器人全局避障最优路径,采用改进人工势场法完成局部避障路径规划,最后采用前馈补偿控制器为动力学方程和最优路径建立动态补偿,根据控制器输出的训练结果,实现巡检机器人避障偏差自动控制。实验结果表明:所提方法避障规划能力及避障运动控制能力均较强,避障偏差控制效果好,具有一定应用价值。

煤矿履带式定向钻机路径规划算法

煤矿履带式定向钻机路径规划过程中存在机身体积约束和实际场景下的行驶效率需求,而常用的A^(*)算法搜索速度慢、冗余节点多,且规划路径贴近障碍物、平滑性较差。提出一种以改进A^(*)算法规划全局路径、融合动态窗口法(DWA)规划局部路径的煤矿履带式定向钻机路径规划算法。考虑定向钻机尺寸影响,在传统A^(*)算法中引入安全扩展策略,即在定向钻机和巷道壁、障碍物之间加入安全距离约束,以提高规划路径的安全性;对传统A^(*)算法的启发函数进行自适应权重优化,同时将父节点的影响加入到启发函数中,以提高全局路径搜索效率;利用障碍物检测原理对经上述改进后的A^(*)算法规划路径剔除冗余节点,并使用分段三次Hermite插值进行二次平滑处理,得到全局最优路径。将改进A^(*)算法与DWA融合,进行煤矿井下定向钻机路径规划。利用Matlab对不同工况环境下定向钻机路径规划算法进行仿真对比分析,结果表明:与Dijkstra算法和传统A^(*)算法相比,改进A^(*)算法在保证安全距离的前提下,加快了搜索速度,搜索时间分别平均减少88.5%和63.2%,且在一定程度上缩短了规划路径的长度,路径更加平滑;改进A^(*)算法与DWA融合算法可有效躲避改进A^(*)算法规划路径上的未知障碍物,路径长度较PRM算法和RRT^(*)算法规划的路径分别平均减小5.5%和2.9%。

动车组前端排雪动态响应及雪阻形成机理

为确定动车组前端雪体与排雪阻力间的作用关系,指导积雪加载下动车组高效排雪设计。运用耦合Drucker−Prager失效准则定义下弹塑性积雪本构和光滑粒子流体力学仿真方法,分析积雪冲击时排障面板动态响应及积雪运动状态规律,研究排雪阻力与积雪深度、雪质特性、运行边界间变化规律。基于质量守恒和达朗贝尔原理构建动车组排雪阻力映射模型,揭示动车组前端雪阻形成机理。研究结果表明:设计运行速度下承受排雪载荷的排障面板处于弹性响应阶段,不会发生塑性破坏;排雪深度为10~410 mm、积雪密度为160~480 kg/m^(3)、行驶速度为80~160 km/h时,排雪阻力与积雪厚度、密度均呈线性正相关关系,与排雪速度呈二次正相关关系;排雪阻力理论模型能够准确预测数值计算结果;相比摩擦阻力和切削阻力,动车组冲击过程中积雪运动状态改变是排雪阻力形成的主要原因。

不确定采摘环境下改进RRT算法的机械臂路径规划研究

由于果蔬采摘环境的不确定性和复杂性,机械臂在复杂环境中完成采摘,其路径规划需考虑实时避障。为实现采摘机械臂在不确定环境下安全采摘,提出一种改进RRT的动态避障算法,以提升机械臂在不确定采摘环境的适应性。针对基本快速扩展随机树算法(Rapidly-exploring Random Trees,RRT)在动态环境下迭代时间长、路径长、适应性差等问题,在RRT算法的基础上,引入目标导向策略,把终点以一定概率作为随机采样点的采样方向,提高算法的迭代效率;引入动态检测机制,对已完成规划的初始路径进行实时检测,使算法适应动态变化的环境。通过仿真分析改进RRT算法,结果表明:改进RRT算法的路径减少16%,迭代时间缩短86.5%;同时,动态检测机制使算法适应动态环境。