【作者】 杨斌；

【导师】 丁永生；

【作者基本信息】 东华大学 ， 控制科学与控制工程， 2016， 博士

【摘要】 随着机器人的使用模式从部件式单元应用向系统式应用的方向发展,机器人的应用已经逐渐地进入了大众的视野,同时群体机器人系统因其能够更好地满足人们的需求而得到了越来越多的关注与研究,相关的研究与应用也如雨后春笋般层出不穷。本论文介绍了群体机器人技术的起源、发展与应用,对领域内的主要研究方向和内容进行了分析和探讨,进而提出基于生物智能算法对群体机器人的控制方法进行优化。首先在菌群算法(Bacterial Chemotaxis)的基础上提出了改进菌群算法进行群体机器人自主目标搜索和围捕,利用自组织群体机器人解决该问题,提出了基于细菌趋化性的群体机器人智能控制算法实现目标搜索和围捕。根据目标区域内群体机器人的随机位置生成一个初始坐标系,再将目标区域划分为Voronoi单元,之后群体机器人在目标区域梯度信息的引导下实现目标搜索和围捕。仿真结果表明了算法的有效性及机器人发生故障时的鲁棒性。与群体机器人分布式控制等常用方法相比,模拟实验的结果表明,这种细菌趋局部最优,而且计算效率高。然后在上述算法的基础上,提出基于动态Voronoi图的群体机器人控制算法,在群体机器人进行任务操作时,根据机器人实时位置动态地划分目标区域,以确保机器人下一步的目标位置为距离自身最近的位置。仿真实验结果表明,算法的提出与实施有效地提高了群体机器人区域覆盖和目标围捕的效率,降低了重复探测率。接着,对于群体机器人控制中出现的带时滞随机连续噪声,给出基于随机基因调控网络(SGRN)和多目标优化的滤波器,仿真实验结果表明,滤波器的使用可以明显地提高在有噪声情况下基于SGRN群体机器人控制系统的收敛速度和运行精度。最后,对论文的全部工作进行了总结,并作出了群体机器人控制方法研究展望。更多还原

【Abstract】 As the development of the robots technology, the robot using mode is changing, from component type unit application to the system application. The use of robots has gradually entered the people’s life. At the same time, swarm robot system got more and more attentions and researches because of its better to meet people’s needs than single ones. Related research and applications have sprung up endlessly in recent years. The history, developments and applications for swarm robots are introduced. Some major subjects in proceed are analyzed and discussed. Firstly, we propose a decentralized control algorithm of swarm robot for target search and trapping inspired by bacteria chemotaxis. First, a local coordinate system is established according to the initial positions of the robots in the target area. Then the target area is divided into Voronoi cells. After the initialization, swarm robots start performing target search and trapping missions driven by the proposed bacteria chemotaxis algorithm under the guidance of the gradient information defined by the target. Simulation results demonstrate the effectiveness of the algorithm and its robustness to unexpected robot failure. Compared with other commonly used methods for distributed control of swarm robots, our simulation results indicate that the bacteria chemotaxis algorithm exhibits less vulnerability to local optimum, and high computational efficiency. Secondly, a dynamic Voronoi-based algorithm to solve the area coverage searching problem in decentralized control of sensors-based swarm robots. In the beginning, local coordinate system is established by initial position and the target area of the swarm robots by BC algorithm. Then the target area is divided into Voronoi cells dynamically by the robots moving. The robots move following the concentration gradient of area by the BC algorithm. Simulation results proved the effectiveness of dynamic Voronoi-based algorithm. At last, a bio-inspired filtering via SGRN is proposed to improve estimation accuracy and robustness for the multi-robot localization system which lacks sufficient information of complete models or the process and with varying measurement noise. The proposed bio-inspired filtering is evolved using a multi-objective optimization algorithm subject to minimizing the absolute value of the error between the desired overall performance index and the actual one, and shorten the settling time. From the dynamics of the robot, the solution existence of the proposed filter is shown while with a low computational complexity to obtain a solution from the proposed filter. The simulation results demonstrate that there is a decrease in scrap and eventually an improvement of the multi-robot localization by using the bio-inspired filtering. Some conclusions and future discussion are given at the end of this paper.更多还原