预选择最小距离指标的冗余机器人动态避障算法

从冗余机器人动态避障的实时性出发,提出了一种预选择最小距离指标.利用方向包围盒OBB(oriented bounding box)进行建模,通过相交测试,在计算实时距离之前剔除安全杆件,并采用局部映射距离计算方法,利用目标点坐标值直接得出实时最小距离.在此基础上,建立与最小距离相关的避障增益和逃逸速度,利用冗余机器人零空间映射矩阵进行避障路径规划.最后,通过仿真证明了利用预选择最小距离指标进行动态避障的有效性和高效性,与传统避障算法相比,整体规划时间减少了15.5%.     

多视点距离图像的对准算法

提出一种多视点距离图像的对准算法.该算法将有拒绝的随机抽样和迭代最近点 (ICP: Iterative Closest Point)算法结合起来,采用粗、精对准时不同的评价函数,利用最小二 乘进行多视点之间运动参数的估计.为了快速进行3D点到物体表面的最近距离和最近点的 计算,采用了物体表面的八叉树样条表示.实验结果表明,该对准算法收敛速度较快,抗噪声 能力较强,并且有较高的对准精度.     

巴氏距离和K-L交换结合的特征选择

巴氏距离(Bhattacharyya Distance)特征选择给出了样本最小错误率上界进的特征．但计算量大。为了减少迭代算法计算时间．文本提出巴氏距离和K-L变换结合的特征选择。首先．对整体样本进行K-L变换．去除变化不大的特征．以降低空间雏数。然后．用迭代方法．进行巴氏距离特征选择。本文中用MNIST手写体数字库的计算表明，该方法能够取得好的效果．比单纯使用K-L变换进行特征选择的最小错误率上界要小得多．同时计算时间大大减少。     

Bézier曲线间最近距离的计算方法

针对Bézier曲线间最近距离计算问题,提出一种简捷、可靠的计算方法.该方法以Bernstein多项式算术运算为工具,建立Bézier曲线间最近距离的计算模型;然后充分利用Bézier曲面的凸包性质和de Casteljau分割算法进行求解.该方法几何意义明确,能有效地避免迭代初始值的选择和非线性方程组的求解,并可进一步推广应用于计算Bézier曲线/曲面间的最近距离.实验结果表明,该方法简捷、可靠且容易实现,与Newton-Raphson方法的融合可进一步提高该方法的运行速度.     

一种新的机器人机构距离误差模型及补偿算法

在标定机器人绝对位置精度和实施误差综合补偿过程中,必然涉及到测量系统坐标系与机器人基础坐标系间的变换.由于这一变换很难精确测定.从而给机器人绝对位置精度标定与误差补偿带来了难以克服的困难.本文首次提出了一种新的机器人机构距离误差计算模型及补偿算法,论证了距离误差同样可以作为机器人绝对位置精度的一种度量.利用该模型和算法对机器人进行误差分析和实施误差综合补偿,可避开上述测量系统与机器人系统间的坐标变换,从而简化了机器人绝对位置精度的标定过程,为提高机器人的绝对位置精度开辟了一个新的简便的途径.     

Bezier曲线间最近距离的计算方法

针对Bezier曲线间最近距离计算问题,提出一种简捷、可靠的计算方法．该方法以Bernstein多项式算术运算为工具,建立Bezier曲线间最近距离的计算模型;然后充分利用Bezier曲面的凸包性质和de Casteljau分割算法进行求解．该方法几何意义明确,能有效地避免迭代初始值的选择和非线性方程组的求解,并可进一步推广应用于计算Bezier曲线／曲面间的最近距离．实验结果表明,该方法简捷、可靠且容易实现,与Newton-Raphson方法的融合可进一步提高该方法的运行速度．     

人机协作视觉手部保护系统设计

针对人机协作中人与机器人共享工作空间时的安全问题,设计了一套人机协作视觉手部保护系统,并搭建相应的验证系统。该系统采用深度学习目标检测算法结合双目视觉技术实现对操作人员手部的识别与定位,同时利用手眼标定将视觉定位后的手部坐标转换到机器人基座坐标系下,通过计算操作人员手部与机器人末端执行器之间的距离,机器人自主执行减速、急停等安全策略。经实验验证:当操作人员在机器人工作空间作业时,通过检测手-末端相对位置关系,可以有效避免人机协作过程中机器人末端执行器与手部发生碰撞,达到了保护操作人员安全的目的。     

空间数据库中最小距离聚集查询及其算法

提出了一种新的距离查询形式-最小距离聚集查询,这种查询计算几个对象集中对象到一个中心对象集中对象的距离和,并返回最小的K个距离和.在空间数据库中,对基于R树索引的数据集给出了基于最近邻居的方法和阈值算法来回答查询.通过大量实验对两种算法进行了比较,结果显示阈值算法具有较好的性能.     

基于GPR和深度强化学习的分层人机协作控制

提出了一种基于高斯过程回归与深度强化学习的分层人机协作控制方法,并以人机协作控制球杆系统为例检验该方法的高效性.主要贡献是:1)在模型未知的情况下,采用深度强化学习算法设计了一种有效的非线性次优控制策略,并将其作为顶层期望控制策略以引导分层人机协作控制过程,解决了传统控制方法无法直接应用于模型未知人机协作场景的问题; 2)针对分层人机协作过程中人未知和随机控制策略带来的不利影响,采用高斯过程回归拟合人体控制策略以建立机器人对人控制行为的认知模型,在减弱该不利影响的同时提升机器人在协作过程中的主动性,从而进一步提升协作效率; 3)利用所得认知模型和期望控制策略设计机器人末端速度的控制律,并通过实验对比验证了所提方法的有效性.     

自愿计算网络中新的距离预测模型

描述了自愿计算环境下主-从模式的并行应用的特点及自愿计算网络的拓扑结构,提出了一种该环境下的网络距离预测模型.该模型处于应用级别上、采用被动监控的方式获取距离预测信息,该网络距离预测模型可以减少测量开销并使预测精度满足实际应用所需.分析了距离预测模型的协议实现与代价评估,最后通过构建自愿计算网络环境,运行并行程序,对该模型进行了测试与性能分析,实验结果验证了该模型的高效性与负载均衡性.     

Deep reinforcement learning-based safe interaction for industrial human-robot collaboration using intrinsic reward function

Aiming at human-robot collaboration in manufacturing, the operator's safety is the primary issue during the manufacturing operations. This paper presents a deep reinforcement learning approach to realize the real-time collision-free motion planning of an industrial robot for human-robot collaboration. Firstly, the safe human-robot collaboration manufacturing problem is formulated into a Markov decision process, and the mathematical expression of the reward function design problem is given. The goal is that the robot can autonomously learn a policy to reduce the accumulated risk and assure the task completion time during human-robot collaboration. To transform our optimization object into a reward function to guide the robot to learn the expected behaviour, a reward function optimizing approach based on the deterministic policy gradient is proposed to learn a parameterized intrinsic reward function. The reward function for the agent to learn the policy is the sum of the intrinsic reward function and the extrinsic reward function. Then, a deep reinforcement learning algorithm intrinsic reward-deep deterministic policy gradient (IRDDPG), which is the combination of the DDPG algorithm and the reward function optimizing approach, is proposed to learn the expected collision avoidance policy. Finally, the proposed algorithm is tested in a simulation environment, and the results show that the industrial robot can learn the expected policy to achieve the safety assurance for industrial human-robot collaboration without missing the original target. Moreover, the reward function optimizing approach can help make up for the designed reward function and improve policy performance.     

一种基于三维建图和虚拟现实的人机交互系统

以提高人机共融水平为目的,以救援机器人为背景,提出并实现基于三维建图和虚拟现实(VR)技术的人机交互系统.在该系统中,救援机器人基于多线激光雷达和惯性测量单元(IMU)实时构建环境的三维点云地图,并将建图结果增量式地表示为3D-NDT地图,实时传输至操作台的虚拟现实系统中可视化;同时,操作人员利用虚拟现实系统的交互设备生成机器人的控制指令,控制机器人运动,构成一个完整的人在回路的人机交互系统.该系统在将机器人环境实时在虚拟现实中可视化的基础上,可以给操作人员以极强的沉浸感,有利于操作人员更直接地理解机器人所处环境.此外,该系统作为一种新的人机交互方式,为提高人与机器人的自然交互水平提供了新思路,对促进人机交互技术的发展具有重要意义.     

Human-robot contactless collaboration with mixed reality interface

A control system based on multiple sensors is proposed for the safe collaboration of a robot with a human. New constrained and contactless human-robot coordinated motion tasks are defined to control the robot end-effector so as to maintain a desired relative position to the human head while pointing at it. Simultaneously, the robot avoids any collision with the operator and with nearby static or dynamic obstacles, based on distance computations performed in the depth space of a RGB-D sensor. The various tasks are organized with priorities and executed under hard joint bounds using the Saturation in the Null Space (SNS) algorithm. A direct human-robot communication is integrated within a mixed reality interface using a stereo camera and an augmented reality system. The proposed system is significant for on-line, collaborative quality assessment phases in a manufacturing process. Various experimental validation scenarios using a 7-dof KUKA LWR4 robot are presented.     

A Human-Robot Collaboration Framework for Improving Ergonomics During Dexterous Operation of Power Tools

In this work, we present a novel control approach to human-robot collaboration that takes into account ergonomic aspects of the human co-worker during power tool operations. The method is primarily based on estimating and reducing the overloading torques in the human joints that are induced by the manipulated external load. The human overloading joint torques are estimated and monitored using a whole-body dynamic state model. The appropriate robot motion that brings the human into the suitable ergonomic working configuration is obtained by an optimisation method that minimises the overloading joint torques. The proposed optimisation process includes several constraints, such as the human arm muscular manipulability and safety of the collaborative task, to achieve a task-relevant optimised configuration. We validated the proposed method by a user study that involved a human-robot collaboration task, where the subjects operated a polishing machine on a part that was brought to them by the collaborative robot. A statistical analysis of ten subjects as an experimental evaluation of the proposed control framework is provided to demonstrate the potential of the proposed control framework in enabling ergonomic and task-optimised human-robot collaboration.     

自由曲线曲面距离迭代算法的研究及其可视化处理

针对几何造型和数控加工的需要，对点到面的投影、曲线曲面距离计算的算法进行了深入的分析，注重线面的几何性质，采用数值方法形成了精巧实用的迭代算法、并在MFC下利用OpenGL通用图形接口，对算法结果进行了可视化处理。     

Human-robot cooperation control for installing heavy construction materials

Previously, ASCI (Automation System for Curtain-wall Installation) which combined with a multi-DOF manipulator to a mini-excavator was developed and applied on construction site. As result, the operation by one operator and more intuitive operation method are proposed to improve ASCI's operation method which need one person with a remote joystick and another operating an excavator. The human-robot cooperative system can cope with various and untypical constructing environment through the real-time interacting with a human, robot and constructing environment simultaneously. The physical power of a robot system helps a human to handle heavy construction materials with relatively scaled-down load. Also, a human can feel and response the force reflected from robot end effecter acting with working environment. This paper presents the feasibility study regarding the application of the proposed human-robot cooperation control for construction robot through experiments on a 2DOF manipulator.     

Human-robot collaboration disassembly planning for end-of-life product disassembly process

The disassembly process is the main step of dealing with End-Of-Life (EOL) products. This process is carried out mostly manually so far. Manual disassembly is not efficient economically and the robotic systems are not reliable in dealing with complex disassembly operations as they have high-level uncertainty. In this research, a disassembly planning method based on human-robot collaboration is proposed. This method employs the flexibility and ability of humans to deal with complex tasks, alongside the repeatability and accuracy of the robot. Besides, to increase the efficiency of the process the components are targeted based on the remanufacturability parameters. First, human-robot collaboration tasks are classified, and using evaluation of components remanufacturability parameters, human-robot collaboration definition and characteristics are defined. To target the right components based on their remanufacturability factors, the PROMETHEE II method is employed to select the components based on Cleanability, Reparability, and Economy. Then, the disassembly process is represented using AND/OR representation and the mathematical model of the process is defined. New optimization parameters for human-robot collaboration are defined and the genetic algorithm was modified to find a near-optimal solution based on the defined parameters. To validate the task classification and allocation, a 6-DOF TECHMAN robot arm is used to test the peg-out-hole disassembly operation as a common disassembly task. The experiments confirm the task classification and allocation method. Finally, an automotive component was selected as a case study to validate the efficiency of the proposed method. The results in comparison with the Particle Swarm algorithm prove the efficiency and reliability of the method. This method produces a higher quality solution for the human-robot collaborative disassembly process.     

栅格地图环境下机器人速度势实时路径规划

针对智能制造工程环境中移动机器人的自动避障问题,提出一种基于栅格地图的移动机器人速度势实时避障路径规划方法。利用栅格法二值化移动机器人的工作场景,从机器人中心出发向不同方向进行栅格搜索。基于障碍物对移动机器人有排斥作用以及目标点对机器人有吸引作用的思想,通过栅格数的累加计算机器人到障碍物之间的实时距离,并以此为参数,考虑障碍物的形状、最小安全距离等因素的影响来建立负的速度增量函数;以机器人当前位置与目标点的实时距离和角度为参数建立正向速度增量函数。进而在机器人运动学模型基础上,定义速度势函数来对移动机器人进行实时速度驱动。通过设置最小速度增量,避免在零势点处的局部极小点问题;通过设立距离阈值,避免在目标点附近速度增量趋于无穷的问题。通过仿真对所提出的算法进行验证。