一种基于混合学习策略的移动机器人路径规划方法

针对未知环境下移动机器人路径规划问题,以操作条件反射学习机制为基础,根据模糊推理系统和学习自动机的原理,提出一种应用于移动机器人导航的混合学习策略.运用仿生的自组织学习方法,通过不断与外界未知环境交互从而使机器人具有自学习和自适应的功能.仿真结果表明,该方法能使机器人学会避障和目标导航任务,与传统的人工势场法相比,能有效地克服局部极小和振荡情况.     

基于自回归模型的移动机器人路径规划

文章研究了全局环境未知且存在动态障碍物情况下的移动机器人路径规划问题;采用全局规划和局部规划相结合的方法,提出了动态未知环境下移动机器人的一种在线实时路径规划方法;该法利用自回归模型来预测动态障碍物的运动轨迹,并把预测位置上的动态障碍物视为是瞬时静态的,然后对该"静态"障碍物进行避碰路径规划;仿真实验结果表明该法有效可行,具有优化性、实时规划性、高度的稳定性和良好的避障能力.     

基于声纳环传感器的机器人避障研究

任何一种移动机器人要实现未知环境自主导航都必须有效而可靠地感知环境信息,而超声波传感器在检测障碍物距离信息方面应用十分广泛。介绍了旅行家II号声纳环传感系统的设计与实现原理,并对声纳的精度进行了测试。在此基础上,提出了移动机器人一种简单避障策略,并运用2种基本避障实验:静态避障和动态避障,验证了该避障策略的正确性和有效性。     

基于学习机制的移动机器人动态场景自适应导航方法

针对在单一学习机制中,移动机器人自主导航一般只适用于静态场景,适应性差的问题,提出一种动态场景自适应导航方法．该方法通过激光测距仪（LRF）获取周围环境的距离信息,在基于增量判别回归（IHDR）算法的单一学习机制导航的基础上,提出了最远距离优先机制的局部避障环节．该导航方法克服了传统导航方法对环境模型的过度依赖,并且本文提出的基于最远距离优先机制的局部避障算法,解决了基于单一学习机制的导航方法对动态场景适应能力不足的问题．本文将动态场景自适应导航方法应用到了MT-R机器人中,与基于单一学习机制的导航方法进行了对比实验,并且运用提出的局部避障算法,对实验中的激光数据进行了算法性能分析．实验结果证实了该方法的可行性,并显示了该方法在动态场景下的良好表现．     

一种混合结构的移动机器人导航控制策略

研究了未知环境下移动机器人实时的导航控制问题.采用分布式系统将反射式行为、反应式行为与慎思规划相结合,设计了移动机器人导航控制策略.根据激光雷达传感器信息设计了基于栅格的实时避障算法和解锁策略.通过慎思规划解决了复杂环境下的局部势能陷阱问题.通过自行研制的移动机器人IMR01的实验验证了导航策略的有效性.     

未知环境下基于Elman网络力控制的移动机器人避障研究

避障控制一直是移动机器人路径规划的难点.提出了一种未知环境下基于神经网络的机器人动态避障方法,同时把混合力/位置控制结构应用到移动机器人的避障控制中.力控制算法是通过在移动机器人和障碍物之间形成虚拟力场,并对其整定,以使它们两者之间能保持期望距离.由于移动机器人的动力学模型和障碍物的不确定性也会对避障控制的性能造成影响,因此采用Elman神经网络来补偿不确定性,同时整定移动机器人和障碍物之间的精确距离.仿真实验表明该动态避障算法是有效的.     

新型势场模型及在实时运动规划中应用

针对复杂环境中移动机器人的实时运动规划问题,在传统人工势场法基础上,以机器人与动态障碍物的相对位置和相对速度为变量,构建了一种新型的位置和速度势场模型,提出了一种新的自适应势场法.通过变参数约束扰动法解决了势场法中的局部极小问题.实验结果表明,该方法能使机器人选择最佳避障策略实现主动避障,快速地摆脱动态障碍物,规划出光滑路径,从而提高避障效率,较好地解决了动态未知环境中的路径规划问题.     

基于深度强化学习和动态窗口法的移动机器人路径规划

为了应对复杂多变的环境并提高移动机器人实时避障能力,提出了一种基于深度强化学习和动态窗口法的融合路径规划方法.首先,通过将机器人的驱动控制直接作用在速度空间来执行路径规划,从而使机器人具备动态窗口特性.然后,设计并训练一个深度Q网络去逼近移动机器人的状态-动作值函数,进而与环境动态地进行交互和试错,实时调整机器人的移动轨迹,最终为机器人找到最优路径.仿真实验结果表明,论文所提方法在自定义RGB(图像像素)图像的复杂环境中能够使移动机器人保持安全适当的速度行驶,找到无碰撞的最优路径,具有较好的鲁棒性.     

基于行为优先与模糊控制的移动机器人导航

由于动态未知环境下自主移动机器人的导航具有较大困难,为实现自主机器人在动态未知环境下的无碰撞运行,文中将行为优先级控制与模糊逻辑控制相结合,提出4种基本行为控制策略:目标寻找、避障、跟踪和解锁.针对'U'型和'V'型障碍物运行解锁问题,提出了行走路径记忆方法,并通过构建虚拟墙来避免机器人再次走入此类区域.仿真实验表明,所提出的控制策略可有效地运用于复杂和未知环境下自主移动机器人的导航,且具有较好的鲁棒性和适应性.     

基于遗传算法的移动机器人动态避障路径规划方法

本文提出了一种基于遗传算法的简单、有效的移动机器人实时动态避障路径规划方法.为利用遗传算法实时、稳定地进行动态路径规划,本文将复杂的二维路径编码问题简化为一维编码问题,并把路边约束、动态避障要求和最短路径要求融合成一个简单的适度函数.仿真实验表明,本文提出的动态路径规划方法可实时、稳定地产生移动机器人运动的最佳局部规划路径,且具有良好的动态避障性能.该方法也可用于智能车辆的自动导航.     

A local obstacle avoidance method for mobile robots in partially known environment

Local obstacle avoidance is a principle capability for mobile robots in unknown or partially known environment. A series of velocity space methods including the curvature velocity method (CVM), the lane curvature method (LCM) and the beam curvature method (BCM) formulate the local obstacle avoidance problem as one of constrained optimization in the velocity space by taking the physical constraints of the environment and the dynamics of the vehicle into account. We present a new local obstacle avoidance approach that combines the prediction model of collision with the improved BCM. Not only does this method inherit the quickness of BCM and the safety of LCM, but also the proposed prediction based BCM (PBCM) can be used to avoid moving obstacles in dynamic environments.     

Improving collision avoidance for mobile robots in partially known environments: the beam curvature method

This paper presents a new approach to obstacle avoidance for mobile robots in cluttered and unknown or partially unknown environments. The method combines a new directional method, called beam method (BM), to improve the performance of a local obstacle avoidance approach called curvature velocity method (CVM). BM calculates the best one-step heading which is used by CVM to obtain the optimal linear and angular velocities. The resulting combined technique is called beam curvature method (BCM).

Different experiments in populated and dynamic environments have proved to be very successful. The method is able to guide the robot safely and efficiently during long time periods. We present some of these results compared with other methods.     

不确定动态环境下移动机器人的完全遍历路径规划

基于生物激励神经网络、滚动窗口和启发式搜索,提出了一种新的完全遍历路径规划方法．该方法用Grossberg的生物神经网络实现移动机器人的局部环境建模,将滚动窗口的概念引入到局部路径规划,由启发式算法决定滚动窗口内的局域路径规划目标．该方法能在不确定动态环境中有效地实现机器人自主避障的完全遍历路径规划．仿真研究证明了该方法的可用性和有效性．     

Reactive navigation in dynamic environment using a multisensorpredictor

A reactive navigation system for an autonomous mobile robot in unstructured dynamic environments is presented. The motion of moving obstacles is estimated for robot motion planning and obstacle avoidance. A multisensor-based obstacle predictor is utilized to obtain obstacle-motion information. Sensory data from a CCD camera and multiple ultrasonic range finders are combined to predict obstacle positions at the next sampling instant. A neural network, which is trained off-line, provides the desired prediction on-line in real time. The predicted obstacle configuration is employed by the proposed virtual force based navigation method to prevent collision with moving obstacles. Simulation results are presented to verify the effectiveness of the proposed navigation system in an environment with multiple mobile robots or moving objects. This system was implemented and tested on an experimental mobile robot at our laboratory. Navigation results in real environment are presented and analyzed.     

基于向量场的移动机器人动态路径规划

由于简洁、高效等优点,人工势场法已应用于自主移动机器人的在线实时路径规划,并受到广泛关注.目前,人工势场法在处理静态环境、动态匀速环境下的路径规划方面已有许多成果,但是,机器人在全变速环境下进行在线实时路径规划时,会出现路径冗余、避碰不及等现象.为此,将目标关于机器人的相对加速度因素引入引力势场函数中;在斥力势场函数的基础上融合避碰预测、减速避障策略;最终,机器人能够避免大量无谓避障,当与障碍物相对速度较大时能提前避障,且快速跟踪到目标.仿真结果验证了所提方法的有效性.     

基于深度强化学习的移动机器人动态路径规划算法

为了在复杂舞台环境下使用移动机器人实现物品搬运或者载人演出,提出了一种基于深度强化学习的动态路径规划算法。首先通过构建全局地图获取移动机器人周围的障碍物信息,将演员和舞台道具分别分类成动态障碍物和静态障碍物。然后建立局部地图,通过LSTM网络编码动态障碍物信息,使用社会注意力机制计算每个动态障碍物的重要性来实现更好的避障效果。通过构建新的奖励函数来实现对动静态障碍物的不同躲避情况。最后通过模仿学习和优先级经验回放技术来提高网络的收敛速度,从而实现在舞台复杂环境下的移动机器人的动态路径规划。实验结果表明,该网络的收敛速度明显提高,在不同障碍物环境下都能够表现出好的动态避障效果。     

Hybrid Potential Field Based Control of Differential Drive Mobile Robots

This paper suggests a new way for nonholonomic mobile robots to navigate in obstacle environments using potential fields based on navigation functions. The proposed strategy is a time-invariant feedback control design with the distinguishing feature that it requires almost no switching compared to alternative methodologies of the same nature. Asymptotic convergence with collision avoidance for the proposed approach is established analytically, and the method is demonstrated on a differential-drive skid steering mobile robot.     

Trajectory time scaling of a mobile robot to avoid dynamic obstacles on the basis of the INLVO

To ensure the collision safety of mobile robots, the velocity of dynamic obstacles should be considered while planning the robot’s trajectory for high-speed navigation tasks. A planning scheme that computes the collision avoidance trajectory by assuming static obstacles may result in obstacle collisions owing to the relative velocities of dynamic obstacles. This article proposes a trajectory time-scaling scheme that considers the velocities of dynamic obstacles. The proposed inverse nonlinear velocity obstacle (INLVO) is used to compute the nonlinear velocity obstacle based on the known trajectory of the mobile robot. The INLVO can be used to obtain the boundary conditions required to avoid a dynamic obstacle. The simulation results showed that the proposed scheme can deal with typical collision states within a short period of time. The proposed scheme is advantageous because it can be applied to conventional trajectory planning schemes without high computational costs. In addition, the proposed scheme for avoiding dynamic obstacles can be used without an accurate prediction of the obstacle trajectories owing to the fast generation of the time-scaling trajectory.     

Evolutionary programming-based univector field navigation methodfor past mobile robots

Most of navigation techniques with obstacle avoidance do not consider the robot orientation at the target position. These techniques deal with the robot position only and are independent of its orientation and velocity. To solve these problems this paper proposes a novel univector field method for fast mobile robot navigation which introduces a normalized two dimensional vector field. The method provides fast moving robots with the desired posture at the target position and obstacle avoidance. To obtain the sub-optimal vector field, a function approximator is used and trained by evolutionary programming. Two kinds of vector fields are trained, one for the final posture acquisition and the other for obstacle avoidance. Computer simulations and real experiments are carried out for a fast moving mobile robot to demonstrate the effectiveness of the proposed scheme.