乒乓球机器人的视觉伺服系统

视觉伺服的乒乓球机器人系统作为典型的"手眼系统",是研究高速视觉感知和快速伺服运动的理想平台,其涉及的高速物体识别跟踪、快速精确轨迹预测及机械臂伺服准确回球等关键技术在工业、军事等领域有广泛的应用前景.本文提出了乒乓球机器人的高速视觉伺服系统实现方法,包括基于特征直方图统计和快速轮廓搜索的目标识别算法,基于模型参数学习和自适应模型调整的物体运动状态估计和轨迹预测算法,及基于轨迹预测的灵巧臂回球规划算法.通过实验验证了各算法的实时性和高效性,并在165cm高的仿人机器人"悟"和"空"上成功实现了双机器人对打和与人对打任务.     

快速连续反应-避障作业环境下的七自由度灵巧臂轨迹规划

人类经长期学习训练后能对高速物体 (如棒球、乒乓球等)具有快速连续反应作业的运动技能, 从深层次上揭示是由于人体在其训练过程中不断学习优选了相应手臂的动作轨迹, 并储存了丰富的经验和知识. 受人体手臂动作此行为机制启发, 本文提出一种 7-DOF灵巧臂快速连续反应-避障作业的轨迹规划方法. 该方法将灵巧臂对高速物体目标作业的轨迹规划问题转化为动作轨迹参数化优选问题, 考虑作业过程中灵巧臂的机构物理约束和障碍约束条件, 以灵巧臂目标可作业度指标构建适应度函数, 采用粒子群优化 (Particle swarm optimization, PSO)方法优选作业轨迹中的冗余参数; 在此基础上 利用灵巧臂动作轨迹参数化优选方法构建相应作业环境下的知识数据库, 实现灵巧臂对高速物体目标的快速连续反应作业. 以仿人机器人乒乓球对弈作业为例, 将该方法应用于 7-DOF灵巧臂乒乓球作业的轨迹规划中. 数值实验及实际对弈试验结果表明, 该方法不仅能使灵巧臂所规划的轨迹 满足灵巧臂机构物理约束与障碍约束条件, 同时能实现灵巧臂对乒乓球体的快速连续反应作业, 验证了该方法的有效性.     

Model Based Motion State Estimation and Trajectory Prediction of Spinning Ball for Ping-Pong Robots using Expectation-Maximization Algorithm

Motion state “Motion state of a ping-pong ball consists of the flying state and spin state.” estimation and trajectory prediction of a spinning ball are two important but challenging issues for both the promotion of the next generation of robotic table tennis systems and the research on motion analysis of spinning-flying objects. Due to the Magnus force acting on the ball, the flying state “Flying state denotes the real-time translational velocity.” and spin state “Spin state denotes the real-time rotational velocity.” are coupled, which makes the accurate estimation of them a huge challenge. In this paper, we first derive the Extended Continuous Motion Model (ECMM) by clustering the trajectories into multiple categories with a K-means algorithm and fitting them respectively using Fourier series. The ECMM can easily adapt to all kinds of trajectories. Based on the ECMM, we propose a novel motion state estimation method using Expectation-Maximization (EM) algorithm, which in result contributes to an accurate trajectory prediction. In this method, the category in ECMM is treated as a latent variable, and the likelihood of motion state is formulated as a Gaussian Mixture Model (GMM) of the differences between the trajectory predictions and observations. The effectiveness and accuracy of the proposed method is verified by offline evaluation using a collected dataset, as well as online evaluation that the humanoid robotic table tennis system “Wu & Kong” successfully hits the high-speed spinning ball.     

Self-Learning for a Humanoid Robotic Ping-Pong Player

Imitating the learning process of a human playing ping-pong is extremely complex. This work proposes a suitable learning strategy. First, an inverse kinematics solution is presented to obtain the smooth joint angles of a redundant anthropomorphic robot arm in order to imitate the paddle motion of a human ping-pong player. As humans instinctively determine which posture is suitable for striking a ball, this work proposes two novel processes: (i) estimating ball states and predicting trajectory using a fuzzy adaptive resonance theory network, and (ii) self-learning the behavior for each strike using a self-organizing map-based reinforcement learning network that imitates human learning behavior. Experimental results demonstrate that the proposed algorithms work effectively when applied to an actual humanoid robot playing ping-pong.     

类人足球机器人场上目标识别算法

在比赛过程中类人足球机器人的视觉系统需要对足球、球门以及对阵双方机器人进行识别. 考虑到算法的快速性及有效性,采用基于颜色信息的算法对球及球门进行识别,通过球及球门的颜色阈值提取图片中球与球门可能的位置,再由球与球门的背景色或面积信息确定球与球门的正确位置. 对双方机器人的识别,首先提取机器人的特征,然后通过在线实时的监督学习方法训练一组级联分类器,通过训练好的分类器对双方机器人进行检测. 实验表明算法能够快速有效地识别场上目标,且算法具有较好的鲁棒性.     

How to localize humanoids with a single camera?

In this paper, we propose a real-time vision-based localization approach for humanoid robots using a single camera as the only sensor. In order to obtain an accurate localization of the robot, we first build an accurate 3D map of the environment. In the map computation process, we use stereo visual SLAM techniques based on non-linear least squares optimization methods (bundle adjustment). Once we have computed a 3D reconstruction of the environment, which comprises of a set of camera poses (keyframes) and a list of 3D points, we learn the visibility of the 3D points by exploiting all the geometric relationships between the camera poses and 3D map points involved in the reconstruction. Finally, we use the prior 3D map and the learned visibility prediction for monocular vision-based localization. Our algorithm is very efficient, easy to implement and more robust and accurate than existing approaches. By means of visibility prediction we predict for a query pose only the highly visible 3D points, thus, speeding up tremendously the data association between 3D map points and perceived 2D features in the image. In this way, we can solve very efficiently the Perspective-n-Point (PnP) problem providing robust and fast vision-based localization. We demonstrate the robustness and accuracy of our approach by showing several vision-based localization experiments with the HRP-2 humanoid robot.     

Policy gradient learning for a humanoid soccer robot

In humanoid robotic soccer, many factors, both at low-level (e.g., vision and motion control) and at high-level (e.g., behaviors and game strategies), determine the quality of the robot performance. In particular, the speed of individual robots, the precision of the trajectory, and the stability of the walking gaits, have a high impact on the success of a team. Consequently, humanoid soccer robots require fine tuning, especially for the basic behaviors. In recent years, machine learning techniques have been used to find optimal parameter sets for various humanoid robot behaviors. However, a drawback of learning techniques is time consumption: a practical learning method for robotic applications must be effective with a small amount of data. In this article, we compare two learning methods for humanoid walking gaits based on the Policy Gradient algorithm. We demonstrate that an extension of the classic Policy Gradient algorithm that takes into account parameter relevance allows for better solutions when only a few experiments are available. The results of our experimental work show the effectiveness of the policy gradient learning method, as well as its higher convergence rate, when the relevance of parameters is taken into account during learning.     

信息融合算法在机器人足球系统中的应用

最优信息融合Kalman滤波算法给出了实时动态环境中线性方差最小的融合估计。采用该算法对机器人足球系统中的小球进行状态估计和预测,并给出了信息融合处理结构和该算法的具体实现步骤。实验结果表明,该算法可以克服单一视觉传感器采集的数据含有较大噪声等局限性,实现了对小球精确的状态估计和预测,具有可行性和优越性,并且在某一机器人视觉传感器出错时,系统仍具有良好的容错性和鲁棒性。     

Recovering ball motion from a single motion-blurred image

Motion blur often affects the ball image in photographs and video frames in many sports such as tennis, table tennis, squash and golf. In this work, we operate on a single calibrated image depicting a moving ball over a known background, and show that motion-blurred ball images, usually unwelcome in computer vision, bear more information than a sharp image. We provide techniques for extracting such information ranging from low-level image processing to 3D reconstruction, and present a number of experiments and possible applications, such as ball localization with speed and direction measurement from a single image, and ball trajectory reconstruction from a single long-exposure photograph.     

腐蚀/膨胀算法在足球机器人视觉系统中的应用

在FIRA MiroSot机器人足球比赛中,视觉系统是获得比赛场上机器人与球位置信息的唯一途径。视觉系统的识别速度、精度直接影响到比赛的胜负。针对传统视觉系统在机器人足球比赛中获取各实体的位置不够准确的问题,提出了一种结合数学形态学中腐蚀/膨胀算法来处理视觉系统中的实时图像,增加足球机器人视觉系统识别的精度的设计方案。实验结果表明,该方案在没有降低比赛中识别速度的前提下,大大地提高了识别精度。     

基于自学习中枢模式发生器的仿人机器人适应性行走控制

为了克服传统中枢模式发生器(Central pattern generator, CPG)关节空间控制方法的复杂性和局限性, 本文基于自学习中枢模式发生器模型, 提出了一套在线调制和融合多传感器信息的仿人机器人环境自适应行走控制方法.算法难点在于如何在机器人的工作空间将自学习CPG用于工作空间轨迹生成, 并使CPG参数直接和步态模式相关联.本文提出了利用自学习CPG来学习和实时生成机器人质心轨迹和脚掌轨迹的方法, 在线调节机器人步长、抬腿高度和步行速度等关键参数.参考生物反射行为, 利用传感反馈信息激发CPG以产生具有环境适应性的工作空间轨迹, 提升行走质量. 控制系统的参数通过优化算法来进一步改善行走性能.相比于传统的CPG关节空间法, 本文所采用的自学习CPG工作空间法不仅极大简化了CPG网络结构而且提高了仿人机器人行走的适应性.最后, 通过仿人机器人坡面适应性行走的仿真和实验, 验证了所提出控制策略的可行性和有效性.     

自主轮式机器人THMR－V的混合模糊逻辑控制

轮式机器人的控制问题是控制研究的关键问题之一，对高速自主导航的轮式机器人，控制器的实时性、精确性和鲁棒性要求很高．在本文中，根据PID控制和模糊逻辑控制的各自优点，将传统的PID控制与模糊逻辑控制结合起来，提出了一种混合模糊逻辑控制算法. 经实验检验，该算法具有很高的实时性、控制精度和鲁棒性，能够满足机器人高速自主导航的需要．     

A machine learning based intelligent vision system for autonomous object detection and recognition

Existing object recognition techniques often rely on human labeled data conducting to severe limitations to design a fully autonomous machine vision system. In this work, we present an intelligent machine vision system able to learn autonomously individual objects present in real environment. This system relies on salient object detection. In its design, we were inspired by early processing stages of human visual system. In this context we suggest a novel fast algorithm for visually salient object detection, robust to real-world illumination conditions. Then we use it to extract salient objects which can be efficiently used for training the machine learning-based object detection and recognition unit of the proposed system. We provide results of our salient object detection algorithm on MSRA Salient Object Database benchmark comparing its quality with other state-of-the-art approaches. The proposed system has been implemented on a humanoid robot, increasing its autonomy in learning and interaction with humans. We report and discuss the obtained results, validating the proposed concepts.     

基于三质心模型的类人机器人射门算法

为了实现快速稳定的射门动作,提出了一种基于三质心模型的类人机器人射门轨迹规划方法。首先,根据三质心模型,得到包含游动腿轨迹和躯干轨迹的零力矩点(ZMP)方程, 采用三次贝塞尔曲线规划游动腿轨迹和ZMP轨迹,根据ZMP方程求解出类人机器人的躯干轨迹;其次,在双腿支撑相根据线性摆模型计算类人机器人的质心轨迹,实现射门姿态的快速调整;最后,在RoboCup 3D仿真平台中应用此算法实现了类人机器人的快速射门动作,并与其他球队的射门动作进行了对比。实验结果表明：应用该算法仅需手工调试即可快速实现稳定的射门动作,射门动作时间有很大减少,可增强机器人足球队的竞争力。     

乒乓球轨迹预测的研究与仿真

机器人打乒乓球时需要对快速运动的乒乓球的轨迹做出准确的预测。首先对乒乓球飞行轨迹预测中采用的运动学模型以及乒乓球碰撞的物理模型进行了分析,其次主要是对乒乓球飞行轨迹预测中采用的运动学模型的算法进行改进,速度反馈调节系数不再使用定值,而是采用动态系数。同时,反弹前的速度是通过对预测的轨迹用最小二乘法进行曲线拟合求导计算出来的。实验结果表明,预测效果很好。最后为了直观地分析预测轨迹,设计了基于OPENGL的3D仿真平台。     

Error Analysis and Effective Adjustment of the Walking-Ready Posture for a Biped Humanoid Robot

A walking control algorithm is generally a mixture of various controllers; it depends on the characteristics of the target system. Simply adopting one part of another researcher's algorithm does not guarantee an improvement in walking performance. However, this paper proposes an effective algorithm that can be easily adopted to other biped humanoid robots; the algorithm enhances the walking performance and stability of the robot merely by adjusting the walking-ready posture. The walking performance of biped humanoid robots is easily affected by an unsuitable walking-ready posture in terms of accuracy and repeatability. More specifically, low accuracy for the walking-ready posture may cause a large difference between an actual biped robot and its mathematical model, and the low repeatability may disturb the evaluation of the performances of balance controllers. Therefore, this paper first discusses the factors that detrimentally affect bipedal walking performance and their phenomena in the walking-ready posture. The necessary conditions for an ideal walking-ready posture are then defined based on static equilibrium and a suitable adjustment algorithm is proposed. Finally, the effectiveness of the algorithm is verified through dynamic computer simulations.     

基于机器视觉的喷涂机器人轨迹规划与涂装质量检测研究综述

随着智能喷涂技术的快速发展,机器视觉在喷涂机器人系统中的研究和应用引起广泛关注,合理的喷涂轨迹能保障油漆厚度均匀、减少漆膜缺陷产生,并且融合涂装质量检测技术形成闭环的喷涂系统.鉴于此,针对机器视觉在喷涂机器人轨迹规划与涂装质量检测中的研究进行综述.首先,对喷涂系统在现代产品制造中的快速发展所面临的机遇、挑战和机器视觉技术进行介绍;然后,综述基于机器视觉技术的喷涂机器人轨迹规划和涂装质量检测的研究成果,对基于机器视觉的喷涂机器人轨迹规划方法,包括待喷涂工件的三维重建、基于点云数据的喷涂轨迹自动规划和基于视觉伺服的喷涂轨迹补偿进行分析和讨论,并重点介绍机器视觉在涂装质量检测中的应用与研究现状,从数据增强和模型选择两个方面,对不同任务中基于深度学习的涂装质量检测算法性能的改善提供潜在解决方案;最后,总结与展望机器视觉技术在喷涂机器人轨迹规划与涂装质量检测中的研究方法和思路,为喷涂系统朝着智能化、柔性化的方向发展提供参考.     

基于计算机视觉的体感输入原型系统研究

体感输入作为一种新兴的人机交互模式,在新一代智能终端中正发挥着越来越重要的作用。特别是基于计算机视觉的体感输入方式,具有无接触、用户体验好等优势。本文设计并实现了一种基于计算机视觉的体感输入原型系统。该系统提供了一种体感输入解决方案,尤其解决了单摄像头环境下手势灵敏度自适应控制的问题,解决了算法处理速度与前端输出需要不匹配时轨迹预测的问题。该系统具有性能良好、成本可控、对现有系统可扩展性强等优点,具有很好的实用价值。     

Integer inverse kinematics method using Fuzzy logic

In this paper, we propose an integer inverse kinematics method for multijoint robot control. The method reduces computational overheads and leads to the development of a simple control system as the use of fuzzy logic enables linguistic modeling of the joint angle. A small humanoid robot is used to confirm via experiment that the method produces the same cycling movements in the robot as those in a human. In addition, we achieve fast information sharing by implementing the all-integer control algorithm in a low-cost, low-power microprocessor. Moreover, we evaluate the ability of this method for trajectory generation and confirm that target trajectories are reproduced well. The computational results of the general inverse kinematics model are compared to those of the integer inverse kinematics model and similar outputs are demonstrated. We show that the integer inverse kinematics model simplifies the control process.     

Humanoid state estimation using a moving horizon estimator

In this research, a new state estimator based on moving horizon estimation theory is suggested for the humanoid robot state estimation. So far, there are almost no studies on the moving horizon estimator (MHE)-based humanoid state estimator. Instead, a large number of humanoid state estimators based on the Kalman filter (KF) have been proposed. However, such estimators cannot guarantee optimality when the system model is nonlinear or when there is a non-Gaussian modeling error. In addition, with KF, it is difficult to incorporate inequality constraints. Since a humanoid is a complex system, its mathematical model is normally nonlinear, and is limited in its ability to characterize the system accurately. Therefore, KF-based humanoid state estimation has unavoidable limitations. To overcome these limitations, we propose a new approach to humanoid state estimation by using a MHE. It can accommodate not only nonlinear systems and constraints, but also it can partially cope with non-Gaussian modeling error. The proposed estimator framework facilitates the use of a simple model, even in the presence of a large modeling error. In addition, it can estimate the humanoid state more accurately than a KF-based estimator. The performance of the proposed approach was verified experimentally.