一种基于力/位混合控制进行插孔作业的策略

插孔作业是一种典型的机器人操作．在工业化机器人装配中存在大量插孔类的作业．由于位置误差的存在，小间隙比插孔作业的成功率并不高．本文从分析插孔失败的原因入手，提出了一种成功率很高的基于力／位混合控制的插孔策略．实验验证了策略的有效性．

Realization of orientation interpolation of 6-axis articulated robot using quaternion

In general, the orientation interpolation of industrial robots has been done based on Euler angle system which can result in singular point （so-called Gimbal Lock）. However, quaternion interpolation has the advantage of natural （specifically smooth） orientation interpolation without Gimbal Lock. This work presents the application of quatemion interpolation, specifically Spherical Linear IntERPolation （SLERP）, to the orientation control of the 6-axis articulated robot （RS2） using LabVIEW and RecurDyn. For the comparison of SLERP with linear Euler interpolation in the view of smooth movement （profile） of joint angles （torques）, the two methods are dynamically simulated on RS2 by using both LabVIEW and RecurDyn. Finally, our original work, specifically the implementation of SLERP and linear Euler interpolation on the actual robot, i.e. RS2, is done using LabVIEW motion control tool kit. The SLERP orientation control is shown to be effective in terms of smooth joint motion and torque when compared to a conventional （linear） Euler interpolation.

Can nurses remain relevant in a technologically advanced future?

Technological breakthroughs occur at an ever-increasing rate thereby revolutionizing human health and wellness care.Technological advancements have drastically changed the structure and organization of the healthcare industry.McKinsey Global Institute estimates that 800 million workers worldwide could be replaced by robots by the year 2030.There is already a robotic revolution happening in healthcare wherein robots have made tasks and procedures more efficient and safer.Locsin and Ito has addressed the threat to nursing practice with human nurses being replaced by humanoid robots.Routine nursing care dictated solely by prescribed procedures and accomplishment of nursing tasks would be best performed by machines.With the future practice of nursing in a technologically advanced future transcending the implementation of nursing actions to achieve predictable outcomes,how can human nurses remain relevant as practitioners of nursing?Nurses should be involved in deciding which aspects of their practice can be delegated to technology.Nurses should oversee the introduction of automated technology and artificial intelligence ensuring their practice to be more about the universal aspects of human care continuing under a novel system.Nursing education and nursing research will change to encompass a differentiated demand for professional nursing practice with,and not for,robots in healthcare.

Relative attribute based incremental learning for image recognition

In this study, we propose an incremental learning approach based on a machine-machine interaction via relative attribute feedbacks that exploit comparative relationships among top level image categories. One machine acts as ＇Student （S）＇ with initially limited information and it endeavors to capture the task domain gradually by questioning its mentor on a pool of unlabeled data. The other machine is ＇Teacher （T）＇ with the implicit knowledge for helping S on learning the class models. T initiates relative attributes as a communication channel by randomly sorting the classes on attribute space in an unsupervised manner. S starts modeling the categories in this intermediate level by using only a limited number of labeled data. Thereafter, it first selects an entropy-based sample from the pool of unlabeled data and triggers the conversation by propagating the selected image with its belief class in a query. Since T already knows the ground truth labels, it not only decides whether the belief is true or false, but it also provides an attribute-based feedback to S in each case without revealing the true label of the query sample if the belief is false. So the number of training data is increased virtually by dropping the falsely predicted sample back into the unlabeled pool. Next, S updates the attribute space which, in fact, has an impact on T＇s future responses, and then the category models are updated concurrently for the next run. We experience the weakly supervised algorithm on the real world datasets of faces and natural scenes in comparison with direct attribute prediction and semi-supervised learning approaches, and a noteworthy performance increase is achieved.

A framework for multi-session RGBD SLAM in low dynamic workspace environment

Mapping in the dynamic environment is an important task for autonomous mobile robots due to the unavoidable changes in the workspace. In this paper, we propose a framework for RGBD SLAM in low dynamic environment, which can maintain a map keeping track of the latest environment. The main model describing the environment is a multi-session pose graph, which evolves over the multiple visits of the robot. The poses in the graph will be pruned when the 3D point scans corresponding to those poses are out of date. When the robot explores the new areas, its poses will be added to the graph. Thus the scans kept in the current graph will always give a map of the latest environment. The changes of the environment are detected by out-of-dated scans identification module through analyzing scans collected at different sessions. Besides, a redundant scans identification module is employed to further reduce the poses with redundant scans in order to keep the total number of poses in the graph with respect to the size of environment. In the experiments, the framework is first tuned and tested on data acquired by a Kinect from laboratory environment. Then the framework is applied to external dataset acquired by a Kinect II from a workspace of an industrial robot in another country, which is blind to the development phase, for further validation of the performance. After this two-step evaluation, the proposed framework is considered to be able to manage the map in date in dynamic or static environment with a noncumulative complexity and acceptable error level.

Immune modelling and programming of a mobile robot demo

An artificial immune system was modelled with self/non-self selection to overcome abnormity in a mobile robot demo. The immune modelling includes the innate immune modelling and the adaptive immune modelling. The self/non-self selection includes detection and recognition, and the self/non-self detection is based on the normal model of the demo. After the detection, the non-self recognition is based on learning unknown non-self for the adaptive immunization. The learning was designed on the neural network or on the learning mechanism from examples. The last step is elimination of all the non-self and failover of the demo. The immunization of the mobile robot demo is programmed with Java to test effectiveness of the approach. Some worms infected the mobile robot demo, and caused the abnormity. The results of the immunization simulations show that the immune program can detect 100% worms, recognize all known Worms and most unknown worms, and eliminate the worms. Moreover, the damaged files of the mobile robot demo can all be repaired through the normal model and immunization. Therefore, the immune modelling of the mobile robot demo is effective and programmable in some anti-worms and abnormity detection applications.

Collision avoidance planning in multi-robot system based on improved artificial potential field and rules

For real-time and distributed features of multi-robot system,the strategy of combining the improved artificial potential field method and the rules based on priority is proposed to study the collision avoidance planning in multi-robot systems. The improved artificial potential field based on simulated annealing algorithm satisfactorily overcomes the drawbacks of traditional artificial potential field method,so that robots can find a local collision-free path in the complex environment. According to the movement vector trail of robots,collisions between robots can be detected,thereby the collision avoidance rules can be obtained. Coordination between robots by the priority based rules improves the real-time property of multi-robot system. The combination of these two methods can help a robot to find a collision-free path from a starting point to the goal quickly in an environment with many obstacles. The feasibility of the proposed method is validated in the VC-based simulated environment.

Simulating the Inverse Kinematic Model of a Robot through Artificial Neural Networks： Complementing the Teaching of Robotics

Teaching robotics necessarily involves the study of the kinematic models of robot manipulators. In turn, the kinematics of a robot manipulator can be described by its forward and reverse models. The inverse kinematic model, which provides the status of the joints according to the desired position for the tool of the robot, is typically taught and described in robotics classes through an algebraic way. However, the algebraic representation of this model is often difficult to obtain. Thus, although it is unquestionable the need for the accurate determination of the inverse kinematic model of a robot, the use of ANNs （artificial neural networks） in the design stage can be very attractive, because it allows us to predict the behavior of the robot before the formal development of its model. In this way, this paper presents a relatively quick way to simulate the inverse kinematic model of a robot, thereby allowing the student to have an overview of the model, coming to identify points that should be corrected, or that can be optimized in the structure of a robot.

Robotic Path Learning with Graphical User Interface

This paper describes an automated path generation method for industrial robots. Based on force control, a robotic subsystem has been developed for path automatic generation or path learning. Using a dummy tool and roughly taught guiding points around a part contour, the robot moves in position and force controlled hybrid mode, following the order of the guiding points and with contact force direction and value predefined. During the motion, robot actual position is recorded by the robot controller. After the motion, the recorded position data is used to generate a robot path program automatically. Robot lead-through may be used in the guiding point teaching. Furthermore, a GUI （graphical user interface） is developed on the teach pedant to guide through the guiding point creation and teaching, path learning, program verification and execution. The development has been incorporated into a robotic machining product option. Combination of the robot path learning function and GUI enhances the interaction between the robot and operator and drastically increases the level of robotic ease-of-use.

Current Status of Intelligent Space

Latest advances in network sensor technology and state of the art of mobile robotics and artificial intelligence research can be applied to develop autonomous and distributed monitoring systems. Intelligent Space （iSpace） is an environmental system, which is able to support human in informative and physical ways. iSpace observing the space with distributed sensors, extracts useful information from the obtained data and provides various services to users. This means that essential functions of iSpace are ＂observation＂, ＂recognition＂ and ＂actuation.＂ In this paper, we focus on the observation function of iSpace. And we describe observation systems to get information of both human and mobile agents in the space to show new results.

Cognitive-affective regulation process for micro-expressions based on Gaussian cloud distribution

In this paper, we explore the process of emotional state transition. And the process is impacted by emotional state of interaction objects. First of all, the cognitive reasoning process and the micro-expressions recognition is the basis of affective computing adjustment process. Secondly, the threshold function and attenuation function are proposed to quantify the emotional changes. In the actual environment, the emotional state of the robot and external stimulus are also quantified as the transferring probability. Finally, the Gaussian cloud distribution is introduced to the Gross model to calculate the emotional transitional probabilities. The experimental results show that the model in human-computer interaction can effectively regulate the emotional states, and can significantly improve the humanoid and intelligent ability of the robot. This model is consistent with experimental and emulational significance of the psychology, and allows the robot to get rid of the mechanical emotional transfer process.

Bionic jumping dynamics of the musculoskeletal leg mechanism for quadruped robots

As the pneumatic artificial muscle （PAM） has flexibility properties similar to biological muscle which is widely used in robotics as one kind of actuators, the bionic mechanism driven by PAMs be- comes a hot spot in robotics. In this paper, a kind of musculoskeletal leg mechanism driven by PAMs is presented, which has three joints driven by four PAMs. The jumping movement is divided into three phases. The forward and inverse kinematics of the leg mechanism in different jumping phases is derived. Considering the ground reaction force between feet and environment, the dynamic in different jumping phases is analyzed by Lagrange method, then the relationship between PAM driving force and the joints angular displacement, angular velocity, angular acceleration during one jumping cycle is obtained, which will lay a foundatiori for the jumping experiment of the musculo- skeletal lez mechanism.

Development of Autonomous Travelling Device for Oil Palm FFB （Fresh Fruit Bunches） Harvester

Robot technology is a very promising technology for agricultural sector, but the existing industrial robot could not deliver the above-mentioned criteria. Industrial robot mainly uses high voltage electrical power, which is not available at field and outdoor operation. The only available and reliable power is a hydraulic from the tractor. The harvester robot consumes the hydraulic power from the tractor and at the same time the tractor can be used as a traveling device for the robot. This paper describes the study on the development of autonomous tractor for the oil palm harvester. The development took considerations on the design of the electro-hydraulic system and the control software for the robot structure to be flexible enough to operate in plantation environment.

Current trends in the development of intelligent unmanned autonomous systems

Intelligent unmanned autonomous systems are some of the most important applications of artificial intelligence （AI）. The development of such systems can significantly promote innovation in AI technologies. This paper introduces the trends in the development of intelligent unmanned autonomous systems by summarizing the main achievements in each technological platform. Furthermore, we classify the relevant technologies into seven areas, including AI technologies, unmanned vehicles, unmanned aerial vehicles, service robots, space robots, marine robots, and unmanned workshops/intelligent plants. Current trends and de- velopments in each area are introduced.

Adaptive Robust Control for an Active Heave Compensation System

Active heave compensation systems are usually employed in offshore and deep-sea operations to reduce the adverse impact of unexpected vessel’s vertical motion on the response of underwater instruments.This paper presents a control strategy for an active heave compensation system consisting of an electro-hydraulic system driven by a double rod actuator,which is subjected to parametric uncertainties and unmeasured environmental disturbances.Adaptive observer and discontinuous projection type updating law with bounded adaption rate are presented firstly to estimate the uncertain system parameters.Then a similar estimation algorithm is designed by using a multiple delayed version of the system to enhance the performance of parameter observation.A reduced order observer is also introduced to estimate unknown wave disturbances.Using the obtained uncertainty information,the resulting control development and stability analysis are implemented based on the Lyapunov’s direct method and back-stepping technique.The proposed controller guarantees the heave compensation error convergent to a bounded neighborhood around the origin.Simulations illustrate the effectiveness of the proposed control system.

An interview with Dr.Raj Reddy on artificial intelligence

Dr.Raj Reddy is the Moza Bint Nasser University Professor of Computer Science and Robotics in the School of Computer Science,Carnegie Mellon University,Pittsburgh,Pennsylvania,USA.He was awarded the ACM Turing Award in 1994.

Three-layer intelligence of planetary exploration wheeled mobile robots:Robint,virtint,and humint

The great success of the Sojourner rover in the Mars Pathfinder mission set off a global upsurge of planetary exploration with autonomous wheeled mobile robots(WMRs),or rovers.Planetary WMRs are among the most intelligent space systems that combine robotic intelligence(robint),virtual intelligence(virtint),and human intelligence(humint) synergetically.This article extends the architecture of the three-layer intelligence stemming from successful Mars rovers and related technologies in order to support the R&D of future tele-operated robotic systems.Double-layer human-machine interfaces are suggested to support the integration of humint from scientists and engineers through supervisory(Mars rovers) or three-dimensional(3D) predictive direct tele-operation(lunar rovers).The concept of multilevel autonomy to realize robint,in particular,the Coupled-Layer Architecture for Robotic Autonomy developed for Mars rovers,is introduced.The challenging issues of intelligent perception(proprioception and exteroception),navigation,and motion control of rovers are discussed,where the terrains' mechanical properties and wheel-terrain interaction mechanics are considered to be key.Double-level virtual simulation architecture to realize virtint is proposed.Key technologies of virtint are summarized:virtual planetary terrain modeling,virtual intelligent rover,and wheel-terrain interaction mechanics.This generalized three-layer intelligence framework is also applicable to other systems that require human intervention,such as space robotic arms,robonauts,unmanned deep-sea vehicles,and rescue robots,particularly when there is considerable time delay.