Learning for joint attention helped by functional development

Cognitive scientists and developmental psychologists have suggested that development in perceptual, motor and memory functions of human infants as well as adaptive evaluation by caregivers facilitate learning for cognitive tasks by infants. This article presents a robotic approach to understanding the mechanism of how learning for joint attention can be helped by such functional development. A robot learns visuomotor mapping needed to achieve joint attention based on evaluations from a caregiver. The caregiver adjusts the criterion for evaluating the robot's performance from easy to difficult as the performance improves. At the same time, the robot also gradually develops its visual function by sharpening input images. Experiments reveal that the adaptive evaluation by the caregiver accelerates the robot's learning and that the visual development in the robot improves the accuracy of joint attention tasks due to its well-structured visuomotor mapping. These results constructively explain what roles synchronized functional development in infants and caregivers play in task learning by infants.     

Acquisition of a symbolic manipulation task model by attention point analysis

As a way of automatic programming of robot behavior, a method for building a symbolic manipulation task model from a demonstration is proposed. The feature of this model is that it explicitly stores the information about the essential parts of a task, i.e. interaction between a hand and an environmental object, or interaction between a grasped object and a target object. Thus, even in different environments, this method reproduces robot motion as similar as possible to that of humans to complete the task while changing the motion during non-essential parts to adapt to the current environment. To automatically determine the essential parts, a method called attention point analysis is proposed; this method searches for the nature of a task using multiple sensors and estimates the parameters to represent the task. A humanoid robot is used to verify the reproduced robot motion based on the generated task model.     

Robust motion control with the consideration algorithm of joint torque saturation for a redundant manipulator

As each joint actuator of a robot manipulator has a limit value of torque, the motion control system should consider the torque saturation. In order to consider the torque saturation in a transient state, this paper proposes a new redundant motion control system using the autonomous consideration algorithm on torque saturation. A Jacobian matrix of a redundant robot manipulator can select the optimal one considering its motion energy in the steady state. When the motion control system carries out fast motion and quick disturbance suppression, a high joint torque is required in a transient state. In the experimental results, under the condition of having a large payload torque and a fast motion reference, the proposed redundant manipulator control realizes the quick robot motion robustly and smoothly.     

Acquisition of a page turning skill for a multifingered hand using reinforcement learning

This paper proposes a method of acquisition of a page turning skill for a multifingered robotic hand using reinforcement learning. The goal of this paper is generation of the manipulation skill of a flexible object without its explicit model and tactile sensation during its control. In this paper, a page turning task is considered as an example of such tasks, where a sheet of paper, generally used in books, is a flexible object. The fingertip trajectories are obtained by reinforcement learning based on simulation. The reward considering a friction condition is given, so that a page turning skill without slip between the finger and the paper is obtained. The validity is confirmed by an experimental system which consists of a 2-d.o.f. manipulator and two 2-d.o.f. fingers.     

Force-guided motions of a 6-d.o.f. industrial robot with a joint space approach

This article deals with the interaction between humans and industrial robots, more specifically with the new design and implementation of an algorithm for force-guided motions of a 6-d.o.f. robot. It may be used to comfortably teach positions without using any teaching pendant or for some assistance tasks. For this purpose, from readings of the force/torque sensor mounted in the robot wrist, the gravity forces and torques first have to be eliminated. To control the robot in joint space, it is then convenient to transform the external force and torque values from Cartesian space into joint space using the manipulator transposed Jacobian. This is why with the present approach the Jacobian matrix of the robot used was calculated. Now, from the computed joint torques, suitable position commands of the robot arm can be generated to obtain the desired behavior. A suggestion for this desired behavior is also included in this article. It is based on the impedance control approach in joint space. The proposed algorithm was implemented with the standard Stäubli RX90B industrial robot.     

The design of natural interaction

This paper addresses the problem of the relationship between humans and technology-enhanced spaces and physical objects (later defined as artifacts). The class of cases here analyzed includes interactive digital signage, information kiosks, home media centers and interactive spaces whose purpose is the communication of a meaning. In this domain, complex interfaces are not needed, as common people interaction with information, content and media is in most cases extremely simple. The topic of specialized interfaces for expert users is not addressed here; the focus is on interfaces for the general public, whose main purpose is the basic fruition of digital information, although such information can be large and complex in its organization. This paper is centered on the need of conceiving computer sensing and information presentation as different aspects of the same interaction design problem, instead of separate research entities.

Ranking systems for evaluation of joint and joint motion stressfulness based on perceived discomforts

This study aims to develop ranking systems for evaluation of the stressfulness of joints and joint motions based on perceived discomforts measured through an experiment. Twenty healthy male subjects participated in the experiment, where discomforts for varying joint motions in the sitting and standing postures were measured using the magnitude estimation. The results showed that the perceived discomforts were affected by the type of joint motions, size of joint motions, and joints. The joints and joint motions were classified into several distinct classes according to perceived stresses. Three ranking systems based on the perceived discomforts were developed, including classification by the joint motions and joints, by types of joint motions, and by the joints only. The ranking systems revealed that while hip and back motions exhibited higher discomfort ratings than any other joint motion, elbow motions were the least stressful of all joint motions. The ranking systems can be used as a valuable design guideline when ergonomically designing or evaluating workplaces, or as a helpful tool for understanding adverse effects of poor working postures.     

Pre-obesity and obesity impacts on passive joint range of motion

Despite the prevalence of pre-obesity and obesity, the physical capabilities of pre-obese/obese individuals are not well documented. As an effort to address this, this study investigated the pre-obesity and obesity impacts on joint range of motion (RoM) for twenty-two body joint motions. A publicly available passive RoM dataset was analysed. Three BMI groups (normal-weight, pre-obese, and obese [Class I]) were statistically compared in joint RoM. The pre-obese and obese groups were found to have significantly smaller RoM means than the normal-weight for elbow flexion and supination, hip extension and flexion, knee flexion and ankle plantar flexion. The pre-obese and obese groups exhibited no significant inter-group mean RoM differences except for knee flexion; for knee flexion, the obese group had significantly smaller RoM means than the pre-obese. The findings would be useful for designing work tasks and products/systems for high BMI individuals and developing digital human models representing differently sized individuals.

Practitioner summary: This study investigated the pre-obesity and obesity impacts on joint range of motion (RoM) by comparing three participant groups: normal-weight; pre-obese and obese. The pre-obese and obese groups had significantly smaller RoM means than the normal-weight for elbow flexion and supination; hip extension and flexion; knee flexion and ankle plantar flexion.

Abbreviations: ANCOVA: Analysis of Covariance; BMI: Body Mass Index; CI: Confidence Interval; RoM: Range of Motion; SPSS: Statistical Package for the Social Sciences     

Acquisition of multi-modal expression of slip through pick-up experiences

To realize adaptive and robust manipulation, a robot should have several sensing modalities and coordinate their outputs to achieve the given task based on underlying constraints in the real environment. This paper discusses acquisition of multi-modal expression of slip consisting of vibration, pressure and vision sensations through pick-up experiences. A sensor network is proposed to acquire the expression, whose learning ability is demonstrated by a real experiment. The applicability of the learned network is also demonstrated by experiments to realize robust and adaptive picking.     

Perceived discomfort functions based on joint moment for various joint motion directions of the upper limb

The aim of the present study was to formulate the relationship between the perceived discomfort and the joint moment ratio for twelve joint motion directions of the upper limb by considering the between-subject variability, and to investigate the effect of joint motion direction. Three approximation models (i.e., linear, exponential, and logistic function models) were compared in terms of the accuracy of predicting the perceived discomfort, and the logistic function was selected because its average error was lowest. The concept of L-R fuzzy number was used to consider the individual variability of perceived discomfort, and a simplified distribution of perceived discomfort was represented. Cluster analysis showed that the twelve discomfort functions formed two clusters: one for elbow flexion and a second for the remaining joint motions. The data show that elbow flexion is more sensitive than other joint motions to increases in the joint moment ratio.     

Acquisition of spatial knowledge through self-directed interaction with a virtual model of a multi-level building: Effects of training and individual differences

Knowledge about complex spatial structures can be acquired through self-directed interaction with virtual models. In the present study, interactive controls enabled flexible exploration of desktop virtual multi-level building models from an allocentric perspective (providing zoom, rotation, and selection of building levels) as well as from the egocentric perspective (providing virtual movement). Short-time training for deliberate exploration were investigated with respect to spatial knowledge acquisition (N = 115, 59 females and 56 males). Four training conditions were included: (1) no training, (2) interaction with a training model with a basic exploration task, (3) cognitive prompts stimulating organisation of spatial information, (4) cognitive and meta-cognitive prompts stimulating planning and controlling the exploration activity. In addition, spatial abilities, real-world spatial strategies and computer game experience were considered as aptitudes. Aptitude variables explained up to 30% of the variance in spatial learning and mediated an effect of sex. Training explained up to 10% of the variance in spatial learning. Qualified training with prompts (conditions 3, 4) did not improve spatial learning compared with training with the basic task (condition 2). Training strongly diminished the role of aptitudes.     

卓越Ⅰ型仿人机器人底层关节控制器的设计

为了实现卓越Ⅰ型机器人在实际环境中稳步行走,设计了底层控制系统中的关节控制器用于控制舵机。选用DSP TMS320F2812作为处理器,充分利用其全数字型,集成度高,体积小,低功耗以及可实现多轴运动的特点,对底层各个关节舵机进行控制,实现既定舵机的运转,以及将舵机的转速和位置等相关传感器的信息反馈给上级控制处理器实现对关节舵机精确的控制,并通过DSP TMS320F2812的两个事件管理器分别同时控制机器人的左右腿的关节舵机单元,提高了机器人在实际行走以及执行任务过程中的实时性、准确性以及稳定性。     

Dynamics of a recurrent neural network acquired through learning a context-based attention task

Selective attention is a very important function for robots acting in the real world. In this function, not only attention itself, but also context extraction and retention are very intelligent processes and are not easily realized. In this article, an attention task is presented in which context information must be extracted from the first pattern presented, and using the context information, a recognition response must be generated from the second pattern presented. An Elman-type recurrent neural network is used to extract and retain the context information. The reinforcement signal that indicates whether the response is correct or not is the only signal given to the system during learning. By this simple learning process, the necessary context information got to be extracted and retained, and then the system changed to generate the correct responses. The function of associative memory was also observed in the feedback-loop in the Elman-type neural network. Furthermore the adaptive formation of basins was examined by varying the learning conditions. This work was presented in part at the 7th International Symposium on Artificial Life and Robotics, Oita, Japan, January 16–18, 2002.     

Enhancing the flow experience of consumers in China through interpersonal interaction in social commerce

Although research on flow experience has recently received much attention, few studies have been published on the perceived interpersonal interaction factors of consumers and their influence in social commerce. In addition, few studies have focused on the impact of interpersonal interaction factors on flow experience. Drawing on the stimulus-organism-response framework, this study examines the impact of interpersonal interaction factors (perceived expertise, similarity, and familiarity) on the formation of flow experience and its subsequent effects on purchase intention in the context of social commerce. We investigate whether the impact of the three interpersonal interaction factors on flow experience differs between young and old users. We conduct a survey and collect 349 responses from users of a social shopping site in China. Our results indicate that interpersonal interaction factors positively relate to flow experience and subsequently influence purchase intention. We also find differences between young and old users in this area.     

Continuity of interaction in nomadic interfaces through migration and dynamic utilization of I/O resources

The concept of ubiquitous computing reflects an infrastructure in which users are engaged in mobile interaction sessions within environments composed of dynamically varying computational resources. In this paradigm, applications are required to continuously follow end users and provide high-quality interaction while migrating among different computing devices and dynamically utilizing the available input/output (I/O) resources of each device. In the context of such interaction scenarios, the principle of continuity, which emphasizes the uninterrupted sequence of dialogue activities, is put forward as a key design goal. This paper presents an application experiment demonstrating interface migration, distributed I/O control, and dynamic I/O reconfiguration. The adopted dialogue design approach is discussed, along with the identified interaction design requirements, from the perspective of dialogue continuity. Finally, the employed software engineering strategy is presented, elaborating on the way dialogue mobility, distribution, and dynamic I/O control have been accomplished. Published online: 11 June 2002     

Increasing teacher candidates' ways of interaction and levels of learning through action research in a blended course

Within the scope of this research, efforts were exerted to increase teacher candidates' interaction ways through action research in a blended teaching profession course in higher education level. Teacher candidates participated in various blended learning activities during a semester-long course, and the problems related to learners' participation in blended learning activities and their interactions during learning process were solved via action decisions. Blended learning activities were developed according to Felder-Silverman Learning Style Model and the lessons learned from previous two pilot studies. In order to conduct the activities; face-to-face and synchronous virtual classroom sessions were combined with asynchronous from discussions and blog. At the beginning of the course, teacher candidates chosed activity sets according to learning style model and throughout a semester-long course, they participated in the activities. Throughout the action research process, the ways of interactions between students, students and instructor, and students and the content were examined as well as the levels of learning accomplished by students during the learning process. This study has revealed indicators pointing to an increase both in students’ interactions and levels of learning during the blended learning process.