动态未知环境下基于相对坐标系的移动机器人实时运动规划

提出了一种简单、新颖的在动态未知环境下的移动机器人运动规划方法 .此方法基于相对坐标系 ,通过传感器信息实时调整机器人的行为来实现规划 .在规划过程中 ,机器人有两种行为 :向目标运动和避碰 ,且避碰行为具有优先权 .机器人两种行为的切换是基于加速度空间的 ,首先解决的是避碰问题 ,而向目标运动是作为避碰的反问题来考虑的 .仿真研究验证了此规划方法的有效性     

一种有指导的轮式机器人全局规划方法*

非完整系统的运动规划是尚未得到充分的重要问题，本文把全局规划方法与局部规划方法结合起来，提出一种包含非完整约束条件的有指导的全局运动规划方法，，并通过仿真验证了其可行性。     

基于蚁群算法的机器人路径规划

移动机器人路径规划是机器人学的一个重要研究领域,栅格法模型是其中一类实时性很强的路径规划模型。该文引入蚁群算法的思想,以点离目标点距离、该点的访问次数和移动方向信息素为启发式因子,建立了一种新型的优化算法。新算法不仅能够较好地对已有算例进行求解,而且对于随机设计的新例子求解效果良好。     

基于协进化路标法的多机器人协作运动规划

针对多关节式机器人协作运动规划方面的难点，在随机路标法的启发下，提出了新颖的协进化路标规划法．这种方法充分利用了协进化算法在优化搜索方面的启发功能和遗传算法对约束条件的处理方法，解决了高维组合C空间的优化搜索问题、静态和动态约束问题和运动轨迹的时间优化问题．并且，通过实验验证了算法的有效性．     

面向三维动画自动生成的路径规划

全过程计算机辅助自动生成动画技术由中科院陆汝钤院士于上世纪90年代提出,这一技术的目标是将适当的故事以受限自然语言的方式输入计算机,由开始一直到动画的生成,每一步都是在计算机的辅助下完成.在动画生成过程中,如何规划运动对象路径是影响动画效果的重要因素.为此,提出基于动画场景规划信息的路径规划方法,该方法通过预定义的路径定性规划语言PADL描述规划需求,然后通过利用扩展的A*算法规划路径并生成路径定量规划语言PCAL描述的路径,最后采用运动图方法合成路径运动动画.实验表明,该方法能有效地自动生成符合指定要求的路径动画.     

加速度约束条件下的非完整移动机器人运动控制

将移动机器人的运动规划与跟踪控制问题合并在一起，对加速度约束条件下的非完整移动机器人运动控制问题进行研究，提出基于贝塞尔曲线的路径规划方法，以满足机器人的非完整约束．在考虑所受加速度约束的条件下，通过规划机器人状态时问轨线的方法实现了时间最优的轨迹规划．基于控制李亚普诺夫函数推导出了轨迹跟踪的控制律．仿真实验结果表明所提出的算法是有效的．     

Flatness-based boundary control of a class of quasilinear parabolic distributed parameter systems

Motion planning and control of a boundary controlled quasilinear parabolic partial differential equation in one spatial variable is considered. The approach relies on a flatness property of the system; namely, that the system solution can be differentially parameterized in terms of a flat output which, in the case considered, is a boundary value. Such a parameterization allows straightforward motion planning and computation of a control law. The approach is based on power series in the spatial variable, and the convergence of these series is ensured by choosing the flat output to be a nonanalytic, smooth function of appropriate Gevrey class.     

基于Leap Motion的手势控制技术初探

最新体感设备Leap Motion的面世提供给用户一种全新的体验, 即通过跟踪探测动态手势可以进行体感游戏、虚拟演奏、凌空绘画等的非接触式人机交互. 文章首先对Leap Motion的技术特点进行介绍, 并对同类型设备进行对比总结, 介绍了Leap Motion的相关应用和发展前景. 文章分析了Leap Motion的原理和技术基础, 然后提出基于Leap Motion的手势控制技术, 最后以一个基于Unity 3D的手势控制虚拟场景中的物品运动的具体实例, 对Leap Motion手势控制技术的实现进行了细节介绍.     

A variational approach to path planning for hyper-redundant manipulators

Motion planning for hyper-redundant manipulators in a complicated and cluttered workspace is a challenging problem. Many of the path planning algorithms, based on cell decomposition or potential field, fail due to the high dimensionality and complex nature of the C-space. Probabilistic roadmap methods (PRM) which have been proven to be successful in high dimensional C-spaces suffer from the drawback of generating paths which involve a lot of redundant motion. In this paper, we propose a path optimizing method to improve a given path in terms of path length and the safety against the collisions, using a variational approach. The capability of variational calculus to optimize a path is demonstrated on a variety of examples. The approach succeeds in providing a good quality path even in high dimensional C-spaces.     

Autonomous control for multi-agent non-uniform spraying

The use of computer controlled swarms of UAVs for crop spraying enables non-uniform coverage of high precision and time efficiency. For this purpose an algorithmic control method for autonomous UAV swarm spraying, based on multi-agent area coverage method Heat Equation Driven Area Coverage (HEDAC), is proposed. Motion control relies on suitable spraying model which enables of multi-agent spraying simulations for arbitrary agent’s trajectories. The HEDAC control method was tested in comparison with conventional (Lawnmower) and state-of-the-art (SMC) path planning methods on three numerical tests: two based on simple geometries and algebraically defined spraying goal densities, and one based on a real-world crop disease map. Additionally, the effects of spraying tool design (number of nozzles and their spraying density) on spraying accuracy were analyzed, with results consistently illustrating the direct causation between tool precision and overall spraying error. The results of the testing have shown HEDAC control to be significantly faster than Lawnmower (approximately 35%–65% less time needed) and SMC (approximately 15%–50% less time needed) in achieving convergence, while producing spraying density of comparable accuracy. Moreover, HEDAC spraying typically mitigates over-spraying by approximately 3%–8% when compared with conventional path planning. In additional tests, it is shown that an implementation of collision avoidance technique for HEDAC motion control provides collision-free UAV swarm spraying. The effect of HEDAC collision avoidance control on the spraying convergence rate and accuracy is practically insignificant. It may be concluded that in real-world application HEDAC controlled UAV spraying swarms are expected to significantly outperform UAVs operating with existing path planning methods.     

室外自主移动机器人AMOR的导航技术

在非结构化环境,移动机器人行驶运动规划和自主导航是非常挑战性的问题。基于实时的动态栅格地图,提出了一个快速的而又实效的轨迹规划算法,实现机器人在室外环境的无碰撞运动导航。AMOR是自主研发的室外运动移动机器人,它在2007年欧洲C-ELROB大赛中赢得了野外自主侦察比赛的冠军。它装备了SICK的激光雷达,用来获取机器人运动前方的障碍物体信息,建立实时动态的环境地图。以A*框架为基础的改造算法,能够在众多的路径中快速地找到最佳的安全行驶路径,实现可靠的自主导航。所有的测试和比赛结果表明所提方案是可行的、有效的。     

Pose Planning for a Mobile Manipulator Based on Joint Motions for Posture Adjustment to End-Effector Error

This paper proposes a motion planning method for a mobile manipulator. In general, humans can grasp an object by various ways which depend on object posture, position and so on. The objective of this paper is to present how to detect the pose of a mobile manipulator under the condition that several ways of grasping are given to the robot. Motion errors and object position errors are considered to detect robot pose in our method because these affect the grasp motion of the robot hand. Coping with these errors, we will propose an effective pose searching method for a mobile manipulator from numerous pose candidates. The performance of the proposed method is illustrated by simulation and experiment.     

Terrain evaluation and its application to path planning for walking machines

Motion planning of walking machines normally contains two aspects: gait planning and body trajectory planning. When generating an optimal body trajectory on natural terrain, the leg movement must be taken into account. Due to the large searching space resulting from the combination of leg movement and terrain conditions, it is quite time consuming to produce an optimal result of body trajectory planning. In this paper, an effective method of body trajectory planning is introduced by virtue of a terrain evaluation that links the terrain conditions with machine mobility. Based on the evaluation, a potential field is constructed for graph searching. Best first planning (BFP) is adopted to search the optimal path. The path generated with the proposed method could offer the best opportunity to place the machine feet moving with a certain gait over a rough terrain. The assumptions and shortages associated with the present work are also discussed.     

基于模糊神经网络的多移动机器人自学习协调系统

研究多移动机器人的运动规划问题．针对机器人模型 未知或不精确以及环境的动态变化,提出一种自学习模糊控制器（FLC）来进行准确的速度 跟踪．首先通过神经网络的学习训练构造FLC,再由再励学习算法来在线调节FLC的输出,以 校正机器人运动状态,实现安全协调避撞．     

A motion capture‐based control‐space approach for walking mannequins

Virtual mannequins need to navigate in order to interact with their environment. Their autonomy to accomplish navigation tasks is ensured by locomotion controllers. Control inputs can be user‐defined or automatically computed to achieve high‐level operations (e.g. obstacle avoidance). This paper presents a locomotion controller based on a motion capture edition technique. Controller inputs are the instantaneous linear and angular velocities of the walk. Our solution works in real time and supports at any time continuous changes of inputs. The controller combines three main components to synthesize locomotion animations in a four‐stage process. First, the Motion Library stores motion capture samples. Motion captures are analysed to compute quantitative characteristics. Second, these characteristics are represented in a linear control space. This geometric representation is appropriate for selecting and weighting three motion samples with respect to the input state. Third, locomotion cycles are synthesized by blending the selected motion samples. Blending is done in the frequency domain. Lastly, successive postures are extracted from the synthesized cycles in order to complete the animation of the moving mannequin. The method is demonstrated in this paper in a locomotion‐planning context. Copyright © 2006 John Wiley & Sons, Ltd.     

Parallel Planning: A New Motion Planning Framework for Autonomous Driving

Motion planning is one of the most significant technologies for autonomous driving. To make motion planning models able to learn from the environment and to deal with emergency situations, a new motion planning framework called as "parallel planning" is proposed in this paper. In order to generate sufficient and various training samples, artificial traffic scenes are firstly constructed based on the knowledge from the reality. A deep planning model which combines a convolutional neural network (CNN) with the Long Short-Term Memory module (LSTM) is developed to make planning decisions in an end-toend mode. This model can learn from both real and artificial traffic scenes and imitate the driving style of human drivers. Moreover, a parallel deep reinforcement learning approach is also presented to improve the robustness of planning model and reduce the error rate. To handle emergency situations, a hybrid generative model including a variational auto-encoder (VAE) and a generative adversarial network (GAN) is utilized to learn from virtual emergencies generated in artificial traffic scenes. While an autonomous vehicle is moving, the hybrid generative model generates multiple video clips in parallel, which correspond to different potential emergency scenarios. Simultaneously, the deep planning model makes planning decisions for both virtual and current real scenes. The final planning decision is determined by analysis of real observations. Leveraging the parallel planning approach, the planner is able to make rational decisions without heavy calculation burden when an emergency occurs.      

基于全方位视觉的多目标跟踪技术

为了快速准确地检测并跟踪多目标对象, 提出了一种基于全方位视觉的多目标对象跟踪方法. 首先采用全方位视觉传感器(ODVS)实时地采集现场360°全景视频图像;接着融合运动历史图像算法(MHI)和运动能量算法(MEI)实现了快速高效的MHoEI(Motion History or Energy Images)自动跟踪算法, 对多目标对象进行检测和跟踪;最后, 本文采用面向对象技术融合目标对象进行匹配跟踪实验结果表明本文提出的方法能较好地跟踪多目标对象, 具有鲁棒性高、运算量小、便于硬件实现、高效等优点.     

基于卷积神经网络的Leap Motion运动数据优化网络

为提高Leap Motion设备的采集精准度,解决自遮挡、采样频率不稳定等设备固有问题,首先,设计了使用Leap Motion和动作捕捉设备的手部多模态同步运动采集方案,采集了日常动作数据集;其次,提出了基于卷积神经网络(convolutional neural network,CNN)的Leap Motion手部运动数据优化方法,使用日常动作数据集训练Leap Motion数据到动作捕捉数据的映射网络;最后,提出手指平面约束,确保网络输出数据保持稳定的手部骨骼结构.通过15名志愿者采集了6类动作共967550帧的同步运动数据集,进行了手指平面约束有效性、动作一致性实验,并与双向循环自编码器(bidirectional recurrent autoencoder,BRA)、双向编解码器(encoder-bidirectional-decoder,EBD)方法进行了精度对比.结果表明,文中方法支持使用Leap Motion获取固定采样频率且近似动捕设备精度的手部运动数据,效果较BRA和EBD更加稳定平滑.将文中方法应用于康复游戏中,明显减少了交互动作识别的错误次数.     

Path planning for autonomous mobile robot navigation with ant colony optimization and fuzzy cost function evaluation

In the Motion Planning research field, heuristic methods have demonstrated to outperform classical approaches gaining popularity in the last 35 years. Several ideas have been proposed to overcome the complex nature of this NP-Complete problem. Ant Colony Optimization algorithms are heuristic methods that have been successfully used to deal with this kind of problems. This paper presents a novel proposal to solve the problem of path planning for mobile robots based on Simple Ant Colony Optimization Meta-Heuristic (SACO-MH). The new method was named SACOdm, where d stands for distance and m for memory. In SACOdm, the decision making process is influenced by the existing distance between the source and target nodes; moreover the ants can remember the visited nodes. The new added features give a speed up around 10 in many cases. The selection of the optimal path relies in the criterion of a Fuzzy Inference System, which is adjusted using a Simple Tuning Algorithm. The path planner application has two operating modes, one is for virtual environments, and the second one works with a real mobile robot using wireless communication. Both operating modes are global planners for plain terrain and support static and dynamic obstacle avoidance.     

基于Leap Motion和S曲线的飞行机器人机械臂控制研究

针对带有多自由度机械臂的飞行机器人,提出基于Leap Motion的控制方法以实现机械臂跟随人体手掌位置姿态运动的功能。采用DH方法建立了机械臂数学模型,给出了将Leap Motion获取的人手运动映射到机械臂末端的推导过程。利用7段S型曲线调速方法近似实现舵机角加速度连续没有突变,减轻了舵机快速响应给飞行器带来的冲击问题。设计制作了实物样机对控制方法的可实现性进行验证测试,在飞行测试中,成功地利用Leap Motion控制远端的机械臂抓取到地面目标。