Kansei and human experience analysis for mobile robot navigation in a ubiquitous environment

Researchers are trying to construct an intelligent system with the help of advance in technology. Many successful milestones have been passed in achieving “humanoid robots” that are little closer to humans, but still not very good. Although the expectation level for the intelligence of these new systems is still unknown, it can be fairly assumed that reaching a goal at least up to the intelligence level of a human being may be a big success. In order to be closer to a human, it is necessary to analyze the behavioral patterns of humans in order to apply those to other artificial intelligence machines. Humans can be described as creatures of emotion. They have feelings toward themselves as well as towards others. These affect all the dealings they perform. Is there anything these feelings can do to robots? Will there be any improvements to robot by having them? This research project discusses the possibility of applying “feelings” to mobile robots in ubiquitous environments.     

基于智能语音的翻译机器人自动化控制系统设计

为提升自动控制效果,加快翻译速率,设计基于智能语音的翻译机器人自动化控制系统。采集外界智能语音信号,利用A/D转换器得到数字信号,启动语音唤醒模块激活翻译机器人,听写模式识别复杂语音信号,命令模式识别简单语音信号,得到语言文本识别结果,通过深度学习关键词检测方法提取关键词作为翻译机器人的自动化控制指令,通过单片机识别自动化控制指令。实验结果表明,该系统可有效采集外界智能语音信号,提取智能语音信号的关键词,完成翻译机器人自动化控制。     

深度学习在智能机器人中的应用研究综述

机器人发展的趋势是人工智能化,深度学习是智能机器人的前沿技术,也是机器学习领域的新课题。深度学习技术被广泛运用于农业、工业、军事、航空等领域,与机器人的有机结合能设计出具有高工作效率、高实时性、高精确度的智能机器人。为了增强智能机器人在各方面的能力,使其更智能化,介绍了深度学习与机器人有关的研究项目与深度学习在机器人中的各种应用,包括室内和室外的场景识别、机器人的工业服务和家庭服务以及多机器人协作等。最后,对深度学习在智能机器人中应用的未来发展、可能面临的机遇和挑战等进行了讨论。     

基于深度强化学习的机器人运动控制研究进展

复杂未知环境下智能感知与自动控制是目前机器人在控制领域的研究热点之一,而新一代人工智能为其实现智能自动化赋予了可能.近年来,在高维连续状态-动作空间中,尝试运用深度强化学习进行机器人运动控制的新兴方法受到了相关研究人员的关注.首先,回顾了深度强化学习的兴起与发展,将用于机器人运动控制的深度强化学习算法分为基于值函数和策略梯度2类,并对各自典型算法及其特点进行了详细介绍;其次,针对仿真至现实之前的学习过程,简要介绍5种常用于深度强化学习的机器人运动控制仿真平台;然后,根据研究类型的不同,综述了目前基于深度强化学习的机器人运动控制方法在自主导航、物体抓取、步态控制、人机协作以及群体协同等5个方面的研究进展;最后,对其未来所面临的挑战以及发展趋势进行了总结与展望.     

Artificial intelligence and robotic assembly

Traditionally, most industrial robots are programmed by teaching. The emergence of robot-level programming languages has improved the programmer's ability to describe and modify the robot moves. However, commercially available robot-level programming languages still fall short of the robot user's need to program complex tasks, and consequently, are not widely used in industry. There is an increasing need for integrating sensors feedback into the robot system to provide better perception and for improving the capacity of the robot to reason and make decisions intelligently in real time.The role of artificial intelligence in programming and controlling robots is discussed. Available robot programming systems including robot-level, object-level, and task-level languages are reviewed. The importance of developing intelligent robots in broadening the scope of flexible automation and opening the door to new robotic applications in space, under water and in harsh environments is outlined. The current development and implementation of programming and control systems for intelligent robots, at McMaster University, are explained. A number of research issues are discussed such as (1) automatic task planning, (2) knowledge representation and use, (3) world modeling, (4) reasoning in automatic assembly planning, and (5) vision monitoring of actions. Examples of geometric, functional, and handling reasoning, as they apply to assembly, are provided. The systems described in this paper are being implemented in the center for flexible manufacturing research and development. Several pieces of hardware are used, including a six-axis articulated robot, a grey-level vision system with a multi-camera, Micro VAX II, and a variety of graphics monitors. The languages available for software development include Common LISP, C, OPS5, VAL II, PASCAL, and FORTRAN 77. The domain of application is currently focused on mechanical assembly.     

一种全自动除雪机器人设计

市场上现有的小型除雪机械智能化、自动化程度低,工作效率慢,结构复杂,综合性能较差,为了更好地优化除雪机械的结构设计,提升工作效率,提高智能化程度,设计了一种新型全自动除雪机器人装置,采用计算机控制、网络RTK实时差分定位等技术实现除雪机器人的全自动导航路线规划,并通过UG三维建模、3D打印等技术对硬件结构进行了系统的运动仿真模拟设计和优化,制造出实物模型,给出了主要的性能参数;经实验测试,一种全自动除雪机器人能够实现全自动路线规划和导航除雪,定位精度可以控制在5厘米以内,自动化程度高,操作灵活,对地形的适应能力强;对除雪机器人的工作效率进行模拟计算得出,相对于市场上现有的小型除雪机械,其工作效率提高了55.52%;由此可得出以下结论:文章全自动除雪机器人设计综合性能较优,智能化的设计能够使除雪机械精准定位,自动规划路线,提高作业效率。     

Control circuits simplification and computer programs design on bipedal robot

This study is expected to provide the best design of bipedal robot, which will help widen the application in the future. A new design system is presented through simplification of the control circuit, and the design of computer programs not only lowers the research barriers of the robots but also decreases the development costs. This design system can be used for various operations of manufacturing processes to make up the shortages of the flexibility for the robots. In recent years, the progress in electronics and the control technology has made the robots useful not only for dangerous and automatic tasks, but also for advanced and friendly people service. Thus, the robots have been used in the factories for automation and towards the general use in regular life. Of all robots, the bipedal robot attracts the most attention for its humanoid outlook, user-friendly design, and artificial intelligence for the human society. Many bipedal robots are developed to satisfy consumers’ needs. The control circuits and the program design are the key issues to make the bipedal robots. In this study, a synchronous robot controller for 31°-axis freedom is developed. The authors also equip a memory in the hardware architecture to store all moving commands of the bipedal robot. In terms of internal programs, the authors develop a human interface that synchronizes the movement of the robot and collects the pace data of the robot. The authors use the statistical method to analyze the data and establish a database of the robot's movement. With the database, one can finally drive the robot to walk and generate the pace design of the bipedal robots.     

Robot accuracy and stiffness—An experimental study

Investigations in robotics have been mainly directed toward the design of robots with a significant degress of intelligence, capable of high adaptability. However, very little research has been done on the design of a robot manipulator as an executive mechanism of the intelligent control system, whose high quality design will allow the sophisticated functions imposed by the control system.

Experimental investigations of robot accuracy show unacceptable results which limit their future wider application. This is the reason why the off-life programming of robots has not been widely used, without which it is impossible to imagine a more significant integrationof robots in the CIM environment.

The object of our study has been to find out the causes of low degreee of robot accuracy, emphasizing the non-corresponding choice of the robot coordinate system, algorithmic and computational errors and drive and transmission element design. Comparison with corresponding results gained from investigations of CNC machine tools cannot be avoided because of the fact that robots are also machines from which accuracy is required. What is missing in robotics today, but which inevitably is required, is the establishment of a standard methodology for testing robot accuracy and other performance.     

Self-reconfigurable robots

In this article, the concept of a cellular robot that is capable of reconfiguring itself is reviewed. This "self-reconfigurable (SR) robot" exemplifies a new trend in robotics, indeed, we can now build various kinds of SR robots with off-the-shelf technologies of processors, actuators, and sensors. These SR robots, based on modern mechatronics, are still not as adaptable as the liquid metal robot in The Terminator 2 but are just as flexible as any conventional robots     

移动机器人网络化环境资源应用中的关键技术

受到软硬件发展瓶颈的制约,机器人本体的能力始终有限,难以应对各种复杂的应用和日益增长的需求．而在机器人周围的环境中,分布着各种各样的资源,却无法为之所用．本文分析了该问题的本质,提出了解决问题所需要的一些关键技术,分析和比较了UPnP、Web服务和Jini三种现有技术,并基于它们提出了初步的解决方案．实验证明,利用该方案,即使低配置的机器人也能够方便地利用环境中的资源,完成各种复杂的任务,从而突破了机器人本体的限制,为扩展机器人的能力开辟了一条新的途径．     

下肢康复机器人的研究进展与临床应用

社会老龄化进程中,由于疾病或事故导致下肢受损的患者越来越多,患者的运动能力康复变得尤为重要.相对于传统的物理康复治疗方法,利用下肢康复机器人进行运功功能康复训练可以准确评估、大幅改善康复效果,提高康复效率,节约医疗康复资源.作为一种涉及康复医学、机械学、控制学、机器人学、计算机学及人工智能等诸多学科的智能仿生机电设备,下肢康复机器人按照患者的康复作业姿态主要分为坐卧式下肢康复机器人和站立式下肢康复机器人.其中坐卧式下肢康复机器人可以分为脚踏板式和外骨骼式机器人,站立式下肢康复机器人分为悬挂减重式和独立可穿戴式机器人.本文调研了国际上应用较为广泛的几类下肢康复机器人的代表性产品,详细分析了下肢康复机器人的主被动自由度、驱动方式、传动关节、感知技术、运动控制策略和训练模式等关键技术,并对各类机器人典型临床测试案例进行了比较和探讨,对下肢康复机器人未来的技术发展做出展望.     

A hardware intelligent processing accelerator for domestic service robots

ABSTRACT

We present a method for implementing hardware intelligent processing accelerator on domestic service robots. These domestic service robots support human life; therefore, they are required to recognize environments using intelligent processing. Moreover, the intelligent processing requires large computational resources. Therefore, standard personal computers (PCs) with robot middleware on the robots do not have enough resources for this intelligent processing. We propose a ‘connective object for middleware to an accelerator (COMTA),’ which is a system that integrates hardware intelligent processing accelerators and robot middleware. Herein, by constructing dedicated architecture digital circuits, field-programmable gate arrays (FPGAs) accelerate intelligent processing. In addition, the system can configure and access applications on hardware accelerators via a robot middleware space; consequently, robotic engineers do not require the knowledge of FPGAs. We conducted an experiment on the proposed system by utilizing a human-following application with image processing, which is commonly applied in the robots. Experimental results demonstrated that the proposed system can be automatically constructed from a single-configuration file on the robot middleware and can execute the application 5.2 times more efficiently than an ordinary PC.     

A modular design of Bayesian networks using expert knowledge: Context-aware home service robot

Recently, demand for service robots increases, and, particularly, one for personal service robots, which requires robot intelligence, will be expected to increase more. Accordingly, studies on intelligent robots are spreading all over the world. In this situation, we attempt to realize context-awareness for home robot while previous robot research focused on image processing, control and low-level context recognition. This paper uses probabilistic modeling for service robots to provide users with high-level context-aware services required in home environment, and proposes a systematic modeling approach for modeling a number of Bayesian networks. The proposed approach supplements uncertain sensor input using Bayesian network modeling and enhances the efficiency in modeling and reasoning processes using modular design based on domain knowledge. We verify the proposed method is useful as measuring the performance of context-aware module and conducting subjective test.     

移动机器人路径规划综述

移动机器人是目前科学技术发展最活跃的领域之一,在工业、农业、医疗等行业广泛应用,还在城市安全、国防和空间探测领域得到更广的应用。要实现机器人在未知环境下自主作业,具备实时、自主、识别高风险区域和风险管理的能力,路径规划是一个重要环节,规划水平的高低,在一定程度上标志着机器人的智能水平,因此研究机器人路径规划对提高机器人的智能化水平、加快工程化应用具有重要的意义。文章重点分别从全局路径规划和局部路径规划角度对机器人路径规划的研究方法进行了分析与总结,同时分析研究了基于仿生学的智能算法的遗传算法、蚁群算法、粒子群算法,最后展望了移动机器人的未来发展趋势。     

机器人感知与控制关键技术及其智能制造应用

智能机器人在服务国家重大需求,引领国民经济发展和保障国防安全中起到重要作用,被誉为“制造业皇冠顶端的明珠”.随着新一轮工业革命的到来,世界主要工业国家都开始加快机器人技术的战略部署.而智能机器人作为智能制造的重要载体,在深入实施制造强国战略,推动制造业的高端化、智能化、绿色化过程中将发挥重要作用.本文从智能机器人的感知与控制等关键技术的视角出发,重点阐述了机器人的三维环境感知、点云配准、位姿估计、任务规划、多机协同、柔顺控制、视觉伺服等共性关键技术的国内外发展现状.然后,以复杂曲面机器人三维测量、复杂部件机器人打磨、机器人力控智装配等机器人智能制造系统为例,阐述了机器人的智能制造的应用关键技术,并介绍了工程机械智能化无人工厂、无菌化机器人制药生产线等典型案例.最后探讨了智能制造机器人的发展趋势和所面临的挑战.     

监控式遥控机器人初探

本文初步探讨了遥控机器人监控的基本概念,得出这样一种认识:监控方式是机器人向智能化发展的一个恰当模式,监控系统是由人的高级智能与机器人的低级智能构成的系统.监控是系统中这两种智能相互作用的过程,人的智能应当能够在机器智能的不同级别上输入,我们研制了一个遥控机器人监控操作器的实验系统,通过实例和实验作了说明.     

A rescue robot system for collecting information designed for ease of use — a proposal of a rescue systems concept

A remote controlled robot for collecting information in disasters, e.g. earthquakes, is one of most effective applications of robots, because it is very dangerous for human beings to locate survivors in collapsed buildings and, in addition, small robots can move into narrow spaces to find survivors. However, previous rescue systems that use robots have a significant problem — a shortage of operators. In catastrophic disasters, in order to save victims, we must explore wide areas within a limited time. Thus, many rescue robots should be employed simultaneously. However, human interfaces of previous rescue robots were complicated, so that well-trained professional operators were needed to operate the robots and, thus, to use many rescue robots, many professional operators were required. However, in such catastrophic disasters it is difficult to get many professional operators together within a short time. In this paper we address the problem and propose a concept of rescue team organization in which professional rescue staff and volunteer staff work together for handling a catastrophic disaster. We point out the necessity for rescue robots which can be operated easily by non-professional volunteer staff. To realize a rescue robot which can be operated easily, we propose a rescue robot system which has a human interface seen in typical, everyday vehicles and a snake-like robot which has mechanical intelligence. We have demonstrated the validity and the effectiveness of the proposed concept by developing a prototype system.     

Research into the application of AI robots in community home leisure interaction

This paper focuses on comprehensive application of artificial intelligence robots for community-based leisure interaction. We propose a multiple-layer perceptron network to design and implement the intelligent interactive home robot system, which includes establishment of an environment map, autonomous navigation, obstacle-avoidance control and human–machine interaction, to complete the positioning and perception functions required by the robot in the home environment. With this system, community residents use an interactive interface to manipulate robots remotely and create an environmental map. In order for the robot to adapt in this changing environment, the robot needs to have a completely autonomous navigation and obstacle-avoidance-control system. In this study, a long-distance obstacle-avoidance fuzzy system and a short-distance anti-fall obstacle-avoidance fuzzy system were used to enable the robot to accommodate unforeseen changes. This technology proved itself capable of navigating a home environment, ensuring that the robot could instantaneously dodge nearby obstacles and correcting the robot’s path of travel. At the same time, it could prevent the robot from falling off a high dropping point and thereby effectively control the robot’s movement trajectory. After combining the above-mentioned multi-sensor and image recognition functions, the intelligent interactive home robot showed that it clearly has the ability to integrate vision, perception and interaction, and we were able to verify that the robot has the necessary adaptability in changing environments and that the design of such interactive robots can be an asset in the home.

     

Design of service robots

According to the International Federation of Robotics (IFR), "a service robot is a robot which operates semi or fully autonomously to perform services useful to the well being of human and equipment, excluding manufacturing operations" [1]. These devices are typically complex systems requiring the input of knowledge from numerous disciplines. The authors have been using different software engineering techniques for the last 15 years, integrating new paradigms in the service robot development process as they emerged. This has made it possible to achieve rapid development of applications and subsequent maintenance. During the early years (1993?1998), our effortswere directed at the development of software for various kinds of teleoperated robots to performmaintenance tasks in nuclear power plants [2]; during a second phase (1999?2006),we built applications for ship-hull cleaning robots [3]. All this time, we have been applying all the possibilities of software engineering, from the use of paradigms for structured and object-based programming in early developments to the adoption of the current model-driven approach [model-driven engineering (MDE)] [4].