Towards a core ontology for robotics and automation

In this paper, we present the current results of the newly formed IEEE-RAS Working Group, named Ontologies for Robotics and Automation. In particular, we introduce a core ontology that encompasses a set of terms commonly used in Robotics and Automation along with the methodology we have adopted. Our work uses ISO/FDIS 8373 standard developed by the ISO/TC184/SC2 Working Group as a reference. This standard defines, in natural language, some generic terms which are common in Robotics and Automation such as robot, robotic device, etc. Furthermore, we discuss the ontology development process employed along with the problems and decisions taken.     

Robot Cloud: Bridging the power of robotics and cloud computing

Cloud computing is shaping the cyber world and evolves as a key computing and service platform for sharing resources including platforms, software applications and everything in the form of services. This is known “X as a Service”. Although it brings our age unparalleled computing ability and economic benefits, the application of cloud computing is still limited currently in the cyberspace due to the cloud services can only reside in cloud instead of our daily life environment. In fact, there are still a plethora of physical position based on-site service demands that cloud computing could help little due to the “cyber limitation”. In this paper, we aim to integrate the cyber world and the physical world by bringing up the idea of “Robot Cloud” to bridge the power of robotics and cloud computing. To make it possible, we design a novel Robot Cloud stack to support our idea and adopt the service-oriented architecture (SOA) to make the functional modules in the Robot Cloud more flexible, extensible and reusable. Then we develop a prototype of Robot Cloud using the popular Google App Engine to demonstrate our design method. Finally, we conduct the simulation experiments with a “robot show” application scenario to evaluate our scheduling policy and identify the effect of different request distributions and robot center solutions.     

Experience-based optimization of universal manipulation strategies for industrial assembly tasks

The trend towards smaller lot sizes and shorter product life cycles requires automation solutions with higher flexibility. Today’s robotic systems often are uneconomical for frequently changing boundary conditions and varying tasks due to high engineering costs needed for a well-defined supply of parts and pallets. At the same time, even small inaccuracies due to shape deviations in parts or pallets often cause high downtime. This work contributes to the robustness of industrial assembly processes with high inaccuracy concurrent to narrow tolerances. Therefore, contact-based manipulation strategies are defined, which are model-free and object-independent and solve common industrial tasks as palletizing, packaging and machine feeding. While the strategies are robust to inaccuracy up to 5 mm/5° due to localization uncertainty or object displacement, they handle usual industrial assembly tolerance of far below 1mm. The necessary flexibility and reusability for new tasks is guaranteed by hierarchical decomposition into atomic sub-strategies. In order to accelerate the execution, the manipulation strategies are customized to each specific task by unsupervised experience-based learning. The flexibility of the manipulation strategies and the progress in cycle time during the execution are shown on common industrial tasks with varying objects, tolerances and inaccuracies.     

Precise and efficient pose estimation of stacked objects for mobile manipulation in industrial robotics challenges

Object manipulation tasks such as picking up, carrying and placing should be executed based on the information of objects which are provided by the perception system. A precise and efficient pose estimation system has been developed to address the requirements and to achieve the objectives for autonomous packaging, specifically picking up of stacked non-rigid objects. For fine pose estimation, a drawing pin shaped kernel and pinhole filtering methods are used on the roughly estimated pose of objects. The system has been applied in a realistic industrial environment as a challenging scenario for the Challenge 2 – Shop Floor Logistics and Manipulation on a mobile manipulator in the context of the European Robotics Challenges (EuRoC) project.     

Digital twin-based industrial cloud robotics: Framework,control approach and implementation

Industrial cloud robotics (ICR) integrates cloud computing with industrial robots (IRs). The capabilities of industrial robots can be encapsulated as cloud services and used for ubiquitous manufacturing. Currently, the digital models for process simulation, path simulation, etc. are encapsulated as cloud services. The digital models in the cloud may not reflect the real state of the physical robotic manufacturing systems due to inaccurate or delayed condition update and therefore result in inaccurate simulation and robotic control. Digital twin can be used to realize fine sensing control of the physical manufacturing systems by a combination of high-fidelity digital model and sensory data. In this paper, we propose a framework of digital twin-based industrial cloud robotics (DTICR) for industrial robotic control and its key methodologies. The DTICR is divided into physical IR, digital IR, robotic control services, and digital twin data. First, the robotic control capabilities are encapsulated as Robot Control as-a-Service (RCaaS) based on manufacturing features and feature-level robotic capability model. Then the available RCaaSs are ranked and parsed. After manufacturing process simulation with digital IR models, RCaaSs are mapped to physical robots for robotic control. The digital IR models are connected to the physical robots and updated by sensory data. A case is implemented to demonstrate the workflow of DTICR. The results show that DTICR is capable to synchronize and merge digital IRs and physical IRs effectively. The bidirectional interaction between digital IRs and physical IRs enables fine sensing control of IRs. The proposed DTICR is also flexible and extensible by using ontology models.     

云机器人的研究进展

将云机器人作为传统分布式网络机器人的自然衍生和发展,并从这一视角分析了云机器人与其他机器人之间的联系与区别.进一步地,详细介绍和分析了云机器人的技术基础,对国内外基于云的机器人服务平台和云机器人相关研究项目与应用进行了综述;在此基础上,分析了一种较为通用的云机器人系统框架模型.最后归纳总结了云机器人亟待解决的关键技术,并对未来云机器人的发展趋势进行了展望.     

Knowledge driven robotics for kitting applications

This article presents a newly developed knowledge methodology/model that was designed to support the IEEE Robotics and Automation Society’s Ontologies for Robotics and Automation Working Group. This methodology/model allows for the creation of systems that demonstrate flexibility, agility, and the ability to be rapidly re-tasked. The methodology/model will be illustrated through a case study in the area of robotic kit building. Through this case study, the knowledge model will be presented, and automatic tools for optimizing the knowledge representation for planning systems and execution systems will be discussed.     

Specifying and optimizing robotic motion for visual quality inspection

Installation or even just modification of robot-supported production and quality inspection is a tedious process that usually requires full-time human expert engagement. The resulting parameters, e.g. robot velocities specified by an expert, are often subjective and produce suboptimal results. In this paper, we propose a new approach for specifying visual inspection trajectories based on CAD models of workpieces to be inspected. The expert involvement is required only to select – in a CAD system – the desired points on the inspection path along which the robot should move the camera. The rest of the approach is fully automatic. From the selected path data, the system computes temporal parametrization of the path, which ensures smoothness of the resulting robot trajectory for visual inspection. We then apply a new learning method for the optimization of robot speed along the specified path. The proposed approach combines iterative learning control and reinforcement learning. It takes a numerical estimate of image quality as input and produces the fastest possible motion that does not result in the degradation of image quality as output. In our experiments, the algorithm achieved up to 53% cycle time reduction from an initial, manually specified motion, without degrading the image quality. We show experimentally that the proposed algorithm achieves better results compared to some other policy learning approaches. The described approach is general and can be used with different types of learning and feedback signals.     

综述：知识系统的V＆V技术

知识库的异常是影响整个知识系统性能的重要因素之一，因此必须对获取的知识进行校验。本文综述了知识库异常检测和验证的相关研究，给出了异常知识的分类及其危害性，分析了知识库验证困难的原因，介绍了用于知识库验证的静态和动态方法，列举了国际上几个著名的知识库验证工具，并对知识库验证的研究进行了展望。     

层次链专家系统方法在工资审批中的应用

文章介绍了将层次链专家系统方法应用于工资审批系统中工资审批计算过程,给出了层次链知识建模、知识表示的基本构架,描述了对属性、产生式规则和基于案例的推理流程,解决了工资计算的复杂性、多变性以及扩充性问题。     

Towards commissioning,resilience and added value of Augmented Reality in robotics: Overcoming technical obstacles to industrial applicability

Augmented Reality (AR) has the potential for facilitating the interaction with robots by enhancing the operator’s spatial understanding as well as providing further cognitive support, e.g. in order to make manual programming processes more efficient and provide on-site simulation. However, we consider major issues that prevent a widespread use of AR towards robotics in industries. Initially, the commissioning of AR devices for robotic applications demands spatial referencing of robots and AR devices. Fiducial markers are a popular artificial aid, but hard to implement in industrial use cases because of additional efforts and a lack of robustness. A further bottleneck for developing AR applications in robotics is the restricted technical and ergonomic maturity of AR devices, e.g. limited local computation and short product life-cycles. Furthermore, benefit through AR in comparison to classic online and offline programming techniques is still unclear for most applicators. In this paper, we present approaches towards a distributed, hardware-agnostic microservice architecture with standard interfaces for an interoperable usage of AR in robotics. Furthermore, we present marker-less pose estimation methods for articulated robot arms as well as mobile robots based on RGB and depth images that serve automatic referencing. Finally, we reveal further application potential of AR in robotics in terms of combining advantages from online and offline programming.     

An innovative decentralized strategy for I-AUVs cooperative manipulation tasks

In the last years, a challenging field of autonomous robotics is represented by cooperative mobile manipulation carried out in different environments (aerial, terrestrial and underwater environment). As regards cooperative manipulation of Intervention-Autonomous Underwater Vehicles (I-AUVs), this application is characterized by a more complex environment of work, compared to the terrestrial or aerial ones, both due to different technological problems, e.g. localization and communication in underwater environment. However, the use of Autonomous Underwater Vehicle (AUV) and I-AUV will necessarily grow up in the future exploration of the sea. Particularly, cooperative I-AUVs represent the natural evolution of single centralized I-AUV because they may be used in various underwater assembly tasks, such as complex underwater structure construction and maintenance (e.g. underwater pipeline and cable transportation can be carried out by multiple cooperative I-AUVs). Furthermore, underwater search and rescue tasks could be more efficient and effective if multiple I-AUVs were used.In this paper, the authors propose an innovative decentralized approach for cooperative mobile manipulation of I-AUVs. This decentralized strategy is based on a different use of potential field method; in particular, a multi-layer control structure is developed to in parallel manage the coordination of the swarm, the guidance and navigation of the I-AUVs and the manipulation task. The main advantage of the potential field method is that less information is necessary: navigation and control problems are reduced to the evaluation of the distance vector among the vehicles, object and obstacles. Moreover, because of the technological problems caused by the underwater environment, the reduction of the transmitted data is one of the keypoints of this architecture. In MATLAB^®-Simulink^®, the authors have simulated a transportation task of a partially known object along a reference trajectory in an unknown environment, where some obstacles are placed. The task is performed by an I-AUV swarm composed of four vehicles, each one provided of a 7 Degrees Of Freedom (DOFs) robotic arm.     

Integrated architecture for industrial robot programming and control

As robot control systems are traditionally closed, it is difficult to add supplementary intelligence. Accordingly, as based on a new notion of user views, a layered system architecture is proposed. Bearing in mind such industrial demands as computing efficiency and simple factory-floor operation, the control layers are parameterized by means of functional operators consisting of pieces of compiled code that can be passed as parameters between the layers. The required interplay between application-specific programs and built-in motion control is thereby efficiently accomplished. The results from experimental evaluation and several case studies suggest the architecture to be very useful also in an industrial context.     

Experiments with service-oriented architectures for industrial robotic cells programming

Integration of equipment in industrial robot cells is to an increasing part involved with interfacing modern Ethernet technologies and low-cost mass produced devices, such as vision systems, laser cameras, force–torque sensors, soft-PLCs, digital pens, pocket-PCs, etc. This scenario enables integrators to offer powerful and smarter solutions, more adapted to small and medium enterprises (SMEs), capable of integrating process knowledge and interface better with humans. Nevertheless, programming all these devices efficiently requires too much specific knowledge about the devices, their hardware architectures and specific programming languages, details about system communication low-level protocols, and other tricky details at the system level. To address these issues, this paper describes and analyses two of the most interesting service-oriented architectures (SOA) available, which exhibit characteristics that are well adapted to industrial robotics cells. To compare, discuss and evaluate their programming features and applicability a test bed was specially designed, and the two SOA are fully implemented to program the test bed. Special focus is given to the way services are specified and to the orchestration tools used to manage system logic. The obtained results show clearly that using integrations schemes based on SOA reduces system integration time and are more adapted to industrial robotic cell system integrators.     

Natural language understanding based on mental image description language <Emphasis Type="Italic">L</Emphasis><Subscript><Emphasis Type="Italic">md</Emphasis></Subscript> and its application to language-centered robot manipulation

The authors have been working on natural language understanding based on the knowledge representation language L _md and its application to robot manipulation by verbal suggestion. The most remarkable feature of L _md is its capability of formalizing spatiotemporal events in good correspondence with human/robotic sensations and actions, which can lead to integrated computation of sensory, motory and conceptual information. This paper describes briefly the process from text to robot action via semantic representation in L _md and the experimental results of robot manipulation driven by verbal suggestion. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

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.     

An open-source multi-DOF articulated robotic educational platform for autonomous object manipulation

This research presents an autonomous robotic framework for academic, vocational and training purpose. The platform is centred on a 6 Degree Of Freedom (DOF) serial robotic arm. The kinematic and dynamic models of the robot have been derived to facilitate controller design. An on-board camera to scan the arm workspace permits autonomous applications development. The sensory system consists of position feedback from each joint of the robot and a force sensor mounted at the arm gripper. External devices can be interfaced with the platform through digital and analog I/O ports of the robot controller. To enhance the learning outcome for beginners, higher level commands have been provided. Advanced users can tailor the platform by exploiting the open-source custom-developed hardware and software architectures. The efficacy of the proposed platform has been demonstrated by implementing two experiments; autonomous sorting of objects and controller design. The proposed platform finds its potential to teach technical courses (like Robotics, Control, Electronics, Image-processing and Computer vision) and to implement and validate advanced algorithms for object manipulation and grasping, trajectory generation, path planning, etc. It can also be employed in an industrial environment to test various strategies prior to their execution on actual manipulators.