A mobile dual-arm manipulation robot system for stocking and disposing of items in a convenience store by using universal vacuum grippers for grasping items

ABSTRACT

A mobile dual-arm robot with universal vacuum grippers (UVG) for performing stocking and disposing tasks was developed. The robot grasps items in the tasks by using UVGs whose surface adapts to the various shapes of the items. A selection algorithm that determines whether the robot should use one or both manipulators to arrange an item was also developed. A ‘reachable-grasp pose’ is defined as a pose in which the robot’s UVG can easily place an item with a target attitude if it grasps the item. By using the selection algorithm, the robot re-grasps the item by adopting the reachable-grasp pose if the two manipulators do not collide when one is in the current grasp pose and the other is in the reachable-grasp pose. The robot system won the third prize of the Future Convenience Store Challenge 2018. In experiments on stocking and disposing tasks, the robot system achieved success rates of 100% for the stocking task and 80% for the disposing task. We believe that the results of this study will help researchers to develop a robot system for not only the stocking and disposing tasks but also other tasks in convenience stores (like customer interaction).     

简易家庭服务机器人系统设计

设计了一种简易的家庭服务机器人系统模型,系统以EasyARM2100单片机为控制核心,带减速装置的直流电机作为小车的动力装置,普通直流伺服电机作为清扫电机,再加以红外光电传感器、超声波传感器和液晶显示电路构成。系统以单片机为中央控制器,通过接收传感器回传的数据确定小车的位置状态,然后,控制小车的工作,由光码盘实现速度测量。利用超声波传感器模块实现测距,并配合软件计算实现避障,从而实现地面清扫,定时催醒等功能。经过各模块联合调试,机器人对位置状态判断准确,并能够改变PWM占空比对电机调速,从而实现机器人转向,各项指标均达到设计要求。     

利用平面单应分解实现服务机器人视觉伺服

智能空间中家庭服务机器人所需完成的主要任务是协助人完成物品的搜寻、定位与传递。而视觉伺服则是完成上述任务的有效手段。搭建了由移动机器人、机械臂、摄像头组成的家庭服务机器人视觉伺服系统,建立了此系统的运动学模型并对安装在机械臂末端执行器上的视觉系统进行了内外参数标定,通过分解世界平面的单应来获取目标物品的位姿参数,利用所获取的位姿参数设计了基于位置的视觉伺服控制律。实验结果表明,使用平面单应分解方法来设计控制律可简单有效地完成家庭物品的视觉伺服任务。     

基于文化算法的家庭服务机器人任务规划

通过对家庭服务机器人任务规划问题进行形式化描述,给出了问题的求解模型,提出了一种改进的文化算法,通过算法中信念空间和种群空间的相互联系和相互促进实现求解。算法采用独特的编码方式,其种群空间采用遗传算法作为进化手段,采用较为独特的信念提取方式构造算法的信念空间并促使其进化。将该算法用于家庭服务机器人大赛的仿真平台上,证明其有效性。     

A review on reinforcement learning for contact-rich robotic manipulation tasks

Research and application of reinforcement learning in robotics for contact-rich manipulation tasks have exploded in recent years. Its ability to cope with unstructured environments and accomplish hard-to-engineer behaviors has led reinforcement learning agents to be increasingly applied in real-life scenarios. However, there is still a long way ahead for reinforcement learning to become a core element in industrial applications. This paper examines the landscape of reinforcement learning and reviews advances in its application in contact-rich tasks from 2017 to the present. The analysis investigates the main research for the most commonly selected tasks for testing reinforcement learning algorithms in both rigid and deformable object manipulation. Additionally, the trends around reinforcement learning associated with serial manipulators are explored as well as the various technological challenges that this machine learning control technique currently presents. Lastly, based on the state-of-the-art and the commonalities among the studies, a framework relating the main concepts of reinforcement learning in contact-rich manipulation tasks is proposed. The final goal of this review is to support the robotics community in future development of systems commanded by reinforcement learning, discuss the main challenges of this technology and suggest future research directions in the domain.     

一种服务机器人家庭全息地图构建方法研究*

提出了一种满足家庭服务机器人环境认知和智能服务需要的融合环境和目标信息的家庭全息地图。设计了局部几何—全局拓扑的全息地图分层表示模型。分析了机器人坐标系、局部环境坐标系和目标的相对关系,给出了机器人局部环境自定位算法和基于坐标变换的服务机器人全息地图构建方法。家庭环境下机器人实物实验表明,基于局部几何—全局拓扑表示的全息地图,服务机器人路径规划和任务执行效率得到有效提升。     

Development of universal vacuum gripper for wall-climbing robot

Task performed at a height, such as wall inspections are one of the dangerous tasks for humans. Thus, robotic technology for safety inspection is required. This research focuses on developing robots to climb vertical walls with flat and uneven surfaces, e.g. concrete, tile and riveted structure. To have wall-climbing capability, climbing robots use vacuum pads, claws, magnets, intermolecular force, and adhesive. However, each of these approaches has disadvantages. To achieve wall climbing on an uneven surface without scratching and staining, we have developed a novel vacuum pad named the Universal Vacuum Gripper (UVG), which is based on the Universal Gripper (UG). The UG is a robot hand using jamming transition of coffee powder inside a balloon to grip uneven material. The UVG is a vacuum pad with a deformable skirt based on the UG. If the skirt shape is deformed in accordance with the contact surface, air leaks can be avoided. Moreover, the deformed skirt can be stiffened, thereby working as a gripper. Here, we evaluate the proposed gripper, having both grasping and adhesion force. We also develop a wall-climbing robot with UVGs, and evaluate its performance on uneven surfaces under real-world conditions.     

Q-bot: heavy object carriage robot for in-house logistics based on universal vacuum gripper

Q-bot is the human-sized carriage robot for lifting heavy weight objects of in-house logistics, such as storehouse and convenience store. The main feature of Q-bot is the adhesion mechanism beneath the foot, called the turnover prevention Universal Vacuum Gripper (in short TP UVG) that holds its body for turnover prevention and self-weight compensation. Turnover prevention is one of the key technologies of in-house logistic robot for effective use of it. Self-weight compensation is another clue for the robot to achieve the labor work in narrow space. TP UVG is achieved both functions by adhering to uneven ground. The other function of Q-bot is multiple objects graspability based on two-sized Universal Vacuum Gripper by dual-armed manipulation. Q-bot also has omnidirectional movability based on mecanum wheels. In this research, we will report on the development of Q-bot and experiments to prevent the robot from falling when it grabs a heavy object while attached to the ground. We also report Q-bot demonstrations of Future Convenience-Store Challenge in the World Robot Summit 2018.     

Modeling development process of skill-based system for human-like manipulation robot

The importance of system integration is widely recognized in robotics. This motivates the application of Model-Based Systems Engineering (MBSE) approaches to improve the development process of robot systems. This paper models a development process to achieve given task goals using a human-like upper body robot based on MBSE approach. For this purpose, we focus on the domain knowledge of tasks and skills in robotics. Since MBSE is a general methodology, there is a lot of flexibility on the way of proceeding with the analysis and design, and how to utilize models there. Using the concept of tasks and skills is helpful for better uderstanding of the development process. Our process is based on three main concepts: (1) stakeholders of User and Developer, (2) coordination between User and Developer using skills as communication interface, (3) extension development. Making the process explicit helps many stakeholders such as robot makers, system integrators, and engineers in various application domains to join the system development. It is also effective for accumulating experiences and work products of the development. In addition, we can expect that better understanding of the engineering process results in the improvement of the process performed by automation tools and humans cooperatively.     

一种家庭安保机器人运动路径规划的蚁群算法

由于家庭居住环境复杂,家庭安保机器人导航问题难于解决。使用传统蚁群算法,家庭安保机器人容易陷入搜索家庭环境局部极值的困境,无法找出在复杂环境下家庭最优的运动路径。因此,家庭安保机器人设计方案引入混沌理论改良局部个体的质量,利用混沌扰动,能够避免家庭机器人陷入搜索家庭环境局部极值的困境,由最初的混沌行为过渡到群体智能行为,使家庭安保机器人找到最优的运动路径。经仿真实验表明,在复杂的家庭环境下,家庭安保机器人也可以安全避障。     

基于卡尔曼滤波的移动机器人运动目标跟踪

提出了一种基于卡尔曼滤波的运动目标快速跟踪算法。针对复杂背景下彩色运动目标跟踪问题,采用基于颜色特征和形状特征相结合的方法进行目标识别。利用卡尔曼滤波器的预测功能,预测运动目标在下一帧中的位置,将图像全局搜索问题转换为局部搜索,提高了系统的实时性。实验结果表明:该算法满足移动机器人运动控制的实时性要求,实现了对运动目标的快速跟踪。     

An algorithm based on bidirectional searching and geometric constrained sampling for automatic manipulation planning in aircraft cable assembly

Robotic manipulation of deformable linear objects has potential application in aircraft cable assembly. However, it is difficult to be implemented in real tasks due to requiring prediction of the object’s deformation and obstacle-free manipulation planning with high efficiency. Aiming at exploring automatic assembly planning for aircraft cables assembly in narrow cabins with obstacles, this paper proposes a novel planning algorithm named RRT-BwC (Bi-direction with Constrain). Firstly, formulation of the question and the manipulation objects are presented with geometric definitions. Then a bi-RRT-tree searching method is developed to design the planner for overcoming obstacles in the high dimensional planning space. The numerical distance between configurations in the cable shape space are defined to measure the demands for their transition in consideration of the manipulation logics. And the sampling zone constrains to the shape configuration nodes are also discussed. Finally, the algorithm presents a valid manipulation sequence for robotic manipulation. The functionality and performance of the improved RRT approach are demonstrated with a simulated real-world problem of aircraft cable assembly, exhibiting computation efficiency promotion.     

基于通用即插即用的服务监控机器人在智能家居中的应用

针对智能家居系统中多种异构电子设备难以与其他设备集成的问题,采用一种通用即插即用(UPnP)方式将服务机器人集成到智能家居系统中,着重介绍了集成后衍生的新服务.首先系统地介绍了UPnP标准和服务机器人Rovio;然后开发一个PC应用程序作为服务机器人和家庭智能网络之间连接的虚拟桥,形成服务机器人为智能家居网络提供服务的控制回路.最后通过一些实际操作来验证UPnP Rovio虚拟桥的性能,旨在探索服务机器人集成到智能家居的可能性和一些衍生的先进服务.通过路径跟踪、垃圾检测等一些先进服务的实际演示验证了通过UPnP方式将服务机器人集成到智能家居网络中的可行性,并展示了智能家居互通后的巨大潜在利益.     

Multi-level offloading decision on efficient object tracking for humanoid robot

Energy efficiency is an important issue for a humanoid robot to track an object, especially robot using battery packs with limited capacity. The tracking algorithm is implemented by many functions, such as walking patterns, balance control, object recognition, path planning, etc. Performing these functions on a remote server can offload the computation on robot for energy efficiency. The offloading needs a proper decision rule in respect to the efficiency and performance. This paper presents a multi-level decision method by classifying the functions into three levels related to stability, real-time performance, and energy efficiency. Based on these levels, the decision rules are designed to offload these functions to track the object efficiently. They are validated by a humanoid robot platform called DARwIn-OP. This method could also be used to minimize the energy consumption on other tasks of a humanoid robot.     

自主移动机器人足球比赛视觉定位方法综述

综述了RoboCup足球赛中全自主移动机器人基于视觉的定位技术,包括机器人自定位和多机器人协作物体定位.介绍了定位技术的发展情况与分类.从机器人环境构建形式的不同以及先验位姿和概率方法的应用与否等方面,系统地分析和比较了各种自定位方法.对于多机器人协作物体定位,阐述了静态方法和动态跟踪方法.总结了定位过程中需要重点研究的传感器模型构建、图像处理、特征匹配以及协作过程涉及的相关问题.最后就视觉定位存在的问题和技术发展趋势进行了讨论.     

Multiobjective evolution of deep learning parameters for robot manipulator object recognition and grasping

Deep Learning (DL) is currently very popular because of its similarity to the hierarchical architecture of human brain with multiple levels of abstraction. DL has many parameters that influence the network performance. In this paper, we introduce a multiobjective evolutionary algorithm (MOEA) to optimize the DBNN parameters subject to the error rate and the network training time as two conflicting objectives. To verify the effectiveness, the proposed method is applied to the robot object recognition and grasping task. We compare the performance of the optimized DBNN model with a) DBNN with arbitrarily selected parameters and b) Deep Belief Network-Deep Neural Network (DBN-DNN). The results show that optimized DL has a superior performance in terms of training time and recognition success rate. In addition, the optimized DBNN model is effective for real-time robotic implementations.     

基于微分几何的蛇形机器人移动与操作统一动力学模型研究

蛇形机器人本体是一种多关节串联机构,可以在各种环境中运动,并且当一端固定时可以实现操作.本文提出一种蛇形机器人移动与操作的统一动力学建模方法,统一蛇形机器人移动状态及操作状态的动力学方程.机器人从移动状态到操作状态的转换意味着机构上的重构,即移动状态无固定基座,而操作状态有固定基座.应用虚设机构法在机构学上统一这两种状态(即构形空间中的嵌入关系),利用指数积公式描述这两种状态的运动学方程.在Riemann流形上建立起蛇形机器人移动和操作的动力学模型,并在对动力学模型中各项计算分析的基础上发现机器人操作动力学方程可直接由移动动力学方程退化得到,同时应用子流形的Gauss公式给出证明.由此在微分几何框架下建立蛇形机器人移动与操作的统一动力学模型.按照几何的观点将蛇形机器人移动与操作动力学模型的统一看作是子流形问题,并赋予几何意义.较单独针对蛇形机器人的一种状态(移动或操作)的动力学模型而言,这种统一的动力学模型能够更深刻地揭示蛇形机器人动力学的特征.     

A unified dynamic model for locomotion and manipulation of a snake-like robot based on differential geometry

A snake-like robot, whose body is a seried-wound articulated mechanism, can move in various environments. In addition, when one end is fixed on a base, the robot can manipulate objects. A method of dynamic modeling for locomotion and manipulation of the snake-like robot is developed in order to unify the dynamic equations of two states. The transformation from locomotion to manipulation is a mechanism reconfiguration, that is, the robot in locomotion has not a fixed base, but it in manipulation ha...     

Reusable robot system for display and disposal tasks at convenience stores based on a SysML model and RT Middleware

Service robots are increasingly being expected to replace human labor; however, in practical settings, these robots have not been fully utilized, partly because the applications that service robots are expected are quite broad. Suitable hardware and software should be developed in order to cope with a wide variety of applications. Ideally, these resources should be developed using common hardware and software platforms are required. Middleware platforms for robotics software, such as ROS and RT Middleware, have already been developed. However, these platforms still face reusability problems due to the inconsistency present in system architectures. Systems are typically assumed to be reused with a given architecture, and if the user-side software architecture is inconsistent with this assumption, the system's reusability suffers. To address this issue, we propose a procedure for optimizing system development using SysML. In this study, our robot system is designed to complete a display disposal tasks in a convenience store; we verify the validity of the robot system using this task. In addition, to verify the reusability of the developed robot system, we employ system functions developed for another task and demonstrate their reuse and operation.