Imitation Control Research Articles

Inverse Reinforcement Learning for Trajectory Imitation Using Static Output Feedback Control.

This article studies the trajectory imitation control problem of linear systems suffering external disturbances and develops a data-driven static output feedback (OPFB) control-based inverse reinforcement learning (RL) approach. An Expert-Learner structure is considered where the learner aims to imitate expert's trajectory. Using only measured expert's and learner's own input and output data, the learner computes the policy of the expert by reconstructing its unknown value function weights and thus, imitates its optimally operating trajectory. Three static OPFB inverse RL algorithms are proposed. The first algorithm is a model-based scheme and serves as basis. The second algorithm is a data-driven method using input-state data. The third algorithm is a data-driven method using only input-output data. The stability, convergence, optimality, and robustness are well analyzed. Finally, simulation experiments are conducted to verify the proposed algorithms.

Robust real-time upper limb motion transfer from human to humanoid for manipulation skill teaching

This paper presents a framework for real-time high-fidelity upper limb motion transfer from a human to a humanoid robot, intending to provide robust support for robots to learn manipulation skills. Robot imitation of human motion is modeled as a simultaneous wrist and elbow tracking problem, where the coordination between multiple imitation objectives and physical constraints receives special attention. First, for the vision-based motion capture system, an efficient preprocessing method is introduced to map the raw human skeletal points analogously or reflectively to the robot’s workspace. Then, the generation of velocity-level imitation motion commands is formalized as a constrained hierarchical quadratic programing (HQP) problem, in which multiple objectives for achieving imitation are organized into a hierarchy. The wrist tracking task is set as the highest priority, followed by the adjustment of the elbow and the minimization of the joint velocity. Physical limits of the robot joints, as well as strict kinematic constraints regarding self-motion, are considered as constraints for the optimization problem. Finally, a hierarchical recurrent neural network (RNN) is designed to solve this HQP problem through its unique neural dynamics. The performance of this network is theoretically guaranteed with high computational efficiency. Quantitative results show that the proposed motion imitation control scheme offers excellent integrated tracking accuracy of [Formula: see text] m and is robust to different arm length ratios. Using this method, the humanoid robot can imitate the upper body motions of a human as accurately as possible under physical constraints, without significant delays.

Research on Holographic Visualization Verification Platform for Construction Machinery Based on Mixed Reality Technology

As “Industry 4.0” progresses, construction machinery is evolving toward large-scale, automation, and integration, resulting in the equipment becoming increasingly sophisticated, and the designs more difficult. Labor costs, transportation, and time will be huge challenges for construction machinery, and mixed reality technology is one of several possible ways to solve this challenge. The research presented in this paper develops a holographic visual verification platform for a digital prototype of construction machinery based on virtual terminal equipment, through investigating the synchronous remote collaboration of multiple terminal devices in a mixed reality scenario. These included semi-physical virtual-real fusion assembly, multi-person real-time voice communication, dynamic loading of MR model based on a cloud server, virtual imitation control, interface design, and human-computer interaction. The effectiveness of this paper’s method is demonstrated through remote collaborative design cases. These included a double drum roller, loader, and milling planer welding production line, as well as tractor modeling review and virtual simulation manipulation of an aerial work platform. The experimental results show that this visual verification platform is a feasible, low-cost and scalable solution, which brings a qualitative breakthrough to the design, research and development, production and other stages in the field of construction machinery.

Designing an Imitation Control System for a Logistics Complex based on New Approaches to Organizing Cybernetic Interaction

The article discusses the general specifics of the organization of automation of industrial and transport systems, provides the subject area of the implementation of management systems in the logistics aspect. The authors consider the creation of a software model for an imitation system for managing logistics production, taking into account the study of the behaviour line and the class of belonging of the user of the system in relation to the spatial aspect of the configuration of placements. The generalization of the Markov chain in the applied aspect is considered in the article. A new paradigm for organizing the distribution line of goods is formalized through the introduction of software algorithms for the classification and optimization of the trajectory of movement. The results of dynamic simulation of the developed system and optimal ratios for predicting the optimal lighting route with adaptive correlation of the distributed lighting system of the route within the framework of the Mesh network are presented.

Motion Similarity Evaluation between Human and a Tri-Co Robot during Real-Time Imitation with a Trajectory Dynamic Time Warping Model.

Precisely imitating human motions in real-time poses a challenge for the robots due to difference in their physical structures. This paper proposes a human–computer interaction method for remotely manipulating life-size humanoid robots with a new metrics for evaluating motion similarity. First, we establish a motion capture system to acquire the operator’s motion data and retarget it to the standard bone model. Secondly, we develop a fast mapping algorithm, by mapping the BVH (BioVision Hierarchy) data collected by the motion capture system to each joint motion angle of the robot to realize the imitated motion control of the humanoid robot. Thirdly, a DTW (Dynamic Time Warping)-based trajectory evaluation method is proposed to quantitatively evaluate the difference between robot trajectory and human motion, and meanwhile, visualization terminals render it more convenient to make comparisons between two different but simultaneous motion systems. We design a complex gesture simulation experiment to verify the feasibility and real-time performance of the control method. The proposed human-in-the-loop imitation control method addresses a prominent non-isostructural retargeting problem between human and robot, enhances robot interaction capability in a more natural way, and improves robot adaptability to uncertain and dynamic environments.

Synchronous Generator Imitation Control and Dynamic Power Sharing for Distributed Power Generation Systems

In this paper, a control method called synchronous generator imitation control (SGIC) as well as a dynamic power sharing for distributed power generation systems (DPGSs) are presented. In order to imitate the behavior of a synchronous generator (SG), the SGIC method for the voltage source converter (VSC) station is proposed. The SGIC includes three loops, namely active power and frequency loop (Pf loop), reactive power and voltage loop (QU loop) as well as inner current loop. The Pf loop is used to emulate the swing equation of the SG. The exciter of the SG is mimicked by the QU loop. The inner current loop is developed for fast current and voltage regulations as well as current limiting. The system stability performances are analyzed through small-signal model, and the effectiveness of the proposed control is validated by PSCAD/EMTDC simulations. The results show that the Pf loop of the VSC emulates the motion equation of the SG rotor very well, which makes the VSC station to have inertia just as a SG; the VSC station can maintain a stable output voltage and regulate the reactive power at the same time; through the dynamic power sharing, precise power control, frequency offset elimination and system stability improvement are achieved; the system has the merits of fault ride-through capability as well as good dynamic performance.

Space Robot Motion Planning Oriented to Autonomous Capture Path

Anthropomorphic robots are an important part of the field of robotics. Controlling the motion of anthropomorphic robots through human movements has important research significance not only for robot control research, but also for robots in national defense, rescue, monitoring, logistics, medical care, elderly care, nursing, and education. There are a wide range of application scenarios in such fields. With the increasing maturity of motion data capture equipment and anthropomorphic robot equipment and the diversity of application environments, traditional anthropomorphic robot control algorithms can no longer meet application requirements. This paper combines Microsoft’s Kinect and Aldebaran’s Nao robot to propose a set of kinematics equation algorithms that fund the capture path to realize the realtime imitation control of human motion by anthropomorphic robots.

The inhibition of automatic imitation: A meta-analysis and synthesis of fMRI studies

Humans copy other people without their conscious awareness, a behaviour known as automatic imitation. Although automatic imitation forms a key part of daily social interactions, we do not copy other people indiscriminately. Instead, we control imitative tendencies by prioritising some actions and inhibiting others. To date, neuroimaging studies investigating the control of automatic imitation have produced inconsistent findings. Some studies suggest that imitation control relies on a domain-specific neural circuit related to social cognition (the theory-of-mind network). In contrast, other studies show engagement of a domain-general neural circuit that is engaged during a diverse range of cognitive control tasks (the multiple demand network). Given the inconsistency of prior findings, in the current paper we avoided problems associated with interpreting individual studies by performing a meta-analysis. To do so, we used a multi-level kernel density analysis to quantitatively identify consistent patterns of activation across functional magnetic resonance imaging studies investigating the control of imitation. Our results show clear and consistent evidence across studies that the control of automatic imitation is guided by brain regions in the multiple demand network including dorsolateral frontoparietal cortex. In contrast, there was only limited evidence that regions in the theory of mind network were engaged. Indeed, medial prefrontal cortex showed no consistent engagement and right temporoparietal junction engagement may reflect spatial rather than imitative control. As such, the current meta-analysis reinforces the role of domain-general control mechanisms and provides limited evidence in support of the role of domain-specific processes in regulating imitative tendencies. Consequently, neurocognitive models of imitation need updating to place more emphasis on domain-general control mechanisms, as well as to consider more complex organisational structures of control, which may involve contributions from multiple cognitive systems.

Social Process and Cohabitation: A Trend of the New Generations

Cohabitation is a phenomenon which has taken place relatively recently in Thailand and is likely to spread and increase gradually in the future. Currently there are very few systematic studies on cohabitation among adolescents in Thailand and information has been obtained mostly from public media such as; newspapers, television programs and the internet. This qualitative and quantitative research aims to study the social process and cohabitating among Thai Adolescents. One thousand three hundred and twenty-eight respondents, who were in years one to four of their studies, were selected to participate, using a stratified random sampling technique. The survey was designed to use questionnaires to gather information from the respondents. In addition, semi-structured interviews were carried out with students, to provide a more in-depth understanding of the issues under study. The conceptual framework for this study was derived from empirical observation and intuition; imitation, social contact and social control. The results of the study revealed that there is a prevalence of cohabitation among Thai Adolescents. The significant factors that influenced the students’ cohabitation are foreign cultures, including American, Japanese and Korean culture (45.12%). Their peers' cohabitation did not play a key role in the respondents wanting to experience cohabitation. The belief that traditional Thai women should reserve themselves for marriage was accepted among respondents (44.75%). Many of the respondents indicated that their families opposed cohabitation. Additionally, sexual behavior of Thai women is limited in the traditional norm. Cohabitation is considered as untraditional and teenagers are dating each other secretly (42.13 %). The information available and institutional education on the prevention of Sexually Transmitted Infections and Unwanted Pregnancies, allow adolescents to better understand the consequences of pre-marital sexual relationships (45.92%).

Functional Specificity and Sex Differences in the Neural Circuits Supporting the Inhibition of Automatic Imitation

Humans show an involuntary tendency to copy other people's actions. Although automatic imitation builds rapport and affiliation between individuals, we do not copy actions indiscriminately. Instead, copying behaviors are guided by a selection mechanism, which inhibits some actions and prioritizes others. To date, the neural underpinnings of the inhibition of automatic imitation and differences between the sexes in imitation control are not well understood. Previous studies involved small sample sizes and low statistical power, which produced mixed findings regarding the involvement of domain-general and domain-specific neural architectures. Here, we used data from Experiment 1 ( N = 28) to perform a power analysis to determine the sample size required for Experiment 2 ( N = 50; 80% power). Using independent functional localizers and an analysis pipeline that bolsters sensitivity, during imitation control we show clear engagement of the multiple-demand network (domain-general), but no sensitivity in the theory-of-mind network (domain-specific). Weaker effects were observed with regard to sex differences, suggesting that there are more similarities than differences between the sexes in terms of the neural systems engaged during imitation control. In summary, neurocognitive models of imitation require revision to reflect that the inhibition of imitation relies to a greater extent on a domain-general selection system rather than a domain-specific system that supports social cognition.

Implementation about Thread and Internet-based Motion Receiving Imitation Controller for Humanoid

In this paper, the method that allows for the combined control scheme of the imitation and autonomous control for a humanoid robot via Internet is presented. Our approach does not use any prior analytical or mathematical robot modeling (e.g. forward or inverse kinematics). All the way from the human operator with encoders to the humanoid joints, many server motors and encoders are connected with single communication path for the imitation control. The key concept for autonomous control is added to let the robot stabilize by utilization of tracing the CoP (center of pressure) and controlling the ankle of the robot’s support foot prior to impact in real time. Thread programming technology is introduced to enhance the control and operation speed up, and explained how to operate joints. The experimental results demonstrate that the humanoid robot can successfully imitate the motions of the operator and can walk stably by both utilizing the CoP and controlling the robot’s ankles.

Brain and psychological mediators of imitation: sociocultural versus physical traits.

The acquisition of cultural beliefs and practices is fundamental to human societies. The psychological and neural mechanisms underlying cultural acquisition, however, are not well understood. Here we used brain imaging to investigate how others' physical and sociocultural attributes may influence imitative learning, a critical component of cultural acquisition. While undergoing fMRI, 17 European American young adults imitated models from three different racial groups performing novel hand gestures. Participants learned that half the models shared their political ideology and half did not. We found that the model's political ideology-a sociocultural characteristic devoid of any physical correlates-was sufficient to influence imitative accuracy, and that this effect was mediated by changes in feelings of similarity to the model. Furthermore, the relationship between the imitative model's political ideology and imitation accuracy was mediated by brain regions associated with imitation and its control, as well as mentalizing. Finally, comparing these new data with those from one of our previous studies revealed that knowledge of a model's political ideology reduces the influence of the model's race on feelings of similarity to the model and imitation accuracy, as well as activity in brain regions typically activated during imitation. Taken together, these findings suggest that (1) others' sociocultural characteristics influence imitative biases more so than their physical attributes, and (2) that neural systems associated with imitation, imitation control, and mentalizing contribute to this cultural learning process.

Application of Imitation Control and Foot Sensors to Bipedal Robot Walking

In this paper a method that allows for the control of a humanoid robot with foot sensors through an imitation control approach is presented. Our approach does not use any prior analytical or mathematical robot modeling (e.g. forward or inverse kinematics). The key concept is to trace the CoP (center of pressure) as fast as possible and to control the ankle of the robot’s support foot prior to impact. The experimental results demonstrate that a humanoid robot can successfully imitate the motions of the operator and can walk stably by utilizing both the CoP and controlling the robot’s ankle.

Implementation of Encoder, Foot Sensor and Walk for Imitation Control of Biped Robot

Programming a biped robot to walk is a challenging issue in robotics. For a stable gait, prior knowledge of the robot’s physical parameters and sophisticated control algorithms are required. Robot walking by imitating a human gait becomes new challenging issues. Using a human motion data to the robot’s walk is convenient even though direct usage of the data usually results in dynamically unstable motion. The purpose of this paper is to develop a new type of encoder and foot sensor for biped imitation control system, which is direct and real-time sensor system. Characteristics of the sensor system are discussed. Finally, the operator’s joint motion can be imitated to the robot by using the new type of wearable encoder sensor. CoP through the developed foot sensor can be obtained and utilized to walk.

Neural systems for preparatory control of imitation.

Humans have an automatic tendency to imitate others. Previous studies on how we control these tendencies have focused on reactive mechanisms, where inhibition of imitation is implemented after seeing an action. This work suggests that reactive control of imitation draws on at least partially specialized mechanisms. Here, we examine preparatory imitation control, where advance information allows control processes to be employed before an action is observed. Drawing on dual route models from the spatial compatibility literature, we compare control processes using biological and non-biological stimuli to determine whether preparatory imitation control recruits specialized neural systems that are similar to those observed in reactive imitation control. Results indicate that preparatory control involves anterior prefrontal, dorsolateral prefrontal, posterior parietal and early visual cortices regardless of whether automatic responses are evoked by biological (imitative) or non-biological stimuli. These results indicate both that preparatory control of imitation uses general mechanisms, and that preparatory control of imitation draws on different neural systems from reactive imitation control. Based on the regions involved, we hypothesize that preparatory control is implemented through top-down attentional biasing of visual processing.

Characteristics of Real-Time Imitation Control for Biped Robot: Communication Time Analysis

To realizing the imitation control of the biped humanoid robot, synchronous question is the core problem. Falling time of the robot and reflecting time of the operator both are the components which to real-time control of the robot. If the robot can walk well in limited time, between the two parts time questions, communication time need us to analyze. Based on the imitation control structure of our robot control system, characters of control model and communication system were studied and conclusions show that synchronous control for a robot by a human is difficult to realize and some other stable control methods need to add in the research.

Characteristics of Real-Time Imitation Control for Biped Robot: Falling Time Analysis

The purpose of this paper is to show that a biped robot can walk by an imitation control. It proposes architecture and system for real-time imitation control of a biped robot. Using this method, the operator can interact with the robot to walk. The operator produces trajectory data necessary to start, stop, walk and redirect the robot. We have to send control commands or new angular position values for to the robot as fast as possible. To get intuition how fast the robot should be controlled, its falling time which as the primary time question is discussed. An inverted pendulum calculation example and the real robot fall down experiment were compared in this paper.

Controlling automatic imitative tendencies: Interactions between mirror neuron and cognitive control systems

Humans have an automatic tendency to imitate others. Although several regions commonly observed in social tasks have been shown to be involved in imitation control, there is little work exploring how these regions interact with one another. We used fMRI and dynamic causal modeling to identify imitation-specific control mechanisms and examine functional interactions between regions. Participants performed a pre-specified action (lifting their index or middle finger) in response to videos depicting the same two actions (biological cues) or dots moving with similar trajectories (non-biological cues). On congruent trials, the stimulus and response were similar (e.g. index finger response to index finger or left side dot stimulus), while on incongruent trials the stimulus and response were dissimilar (e.g. index finger response to middle finger or right side dot stimulus). Reaction times were slower on incongruent compared to congruent trials for both biological and non-biological stimuli, replicating previous findings that suggest the automatic imitative or spatially compatible (congruent) response must be controlled on incongruent trials. Neural correlates of the congruency effects were different depending on the cue type. The medial prefrontal cortex, anterior cingulate, inferior frontal gyrus pars opercularis (IFGpo) and the left anterior insula were involved specifically in controlling imitation. In addition, the IFGpo was also more active for biological compared to non-biological stimuli, suggesting that the region represents the frontal node of the human mirror neuron system (MNS). Effective connectivity analysis exploring the interactions between these regions, suggests a role for the mPFC and ACC in imitative conflict detection and the anterior insula in conflict resolution processes, which may occur through interactions with the frontal node of the MNS. We suggest an extension of the previous models of imitation control involving interactions between imitation-specific and general cognitive control mechanisms.

Study on similarity imitation constraints of biped walking for humanoid robot

Similarity motion imitation on humanoid robot from the captured tracks of human actor has attracted more and more attention. We described the basic system of similarity imitation that was formed by three modules - the image capture, imitation control and imitation on real robot. The similarity degree was analysed by the differences of angle, angular velocity and acceleration between human actor and robot, and the function of spatio-temporal harmony for simultaneous actions was provided as well. In order to make the humanoid robot walk like the human actor by following the captured tracks, the constraints of kinematics and zero moment point ZMP shall be considered. We proposed a method of similarity imitation constraint on biped walking for humanoid robot. The walking process was divided into three phases and the corresponding kinematics and ZMP constraints were set on them. Besides, the trigonometric functions were used to connect the consecutive phases to make the track smooth, and the constraint of avoiding singular pose was also accomplished by making the knee flexible. We applied the constraints of similarity imitation and reached the similar biped walking on the humanoid robot Aldebaran Nao.

A DDS‐based middleware for quality‐of‐service and high‐performance networked robotics

SUMMARYSocial robots must adapt to dynamic environments, human interaction partners and challenging new stringent tasks. Their inner software is usually distributed and should be designed and deployed carefully because slight changes in the robot's requirements can have an important impact not only on the existing source code but also on the resulting performance at run‐time. This paper proposes an implementation of this inner software using a new lightweight middleware for networked robotics called Nerve. The principal novelty this middleware has with respect to other state‐of‐the‐art approaches is that it guarantees both scalability and QoS, which are key requirements for real‐time robotics software. The benefits of Nerve have been proved through its use in two key components of the cognitive system of a social robot: (i) the visual attention mechanism, used to extract relevant data from perceived images; and (ii) a robot learning by imitation control architecture that allows the social robot to be taught by people using natural demonstrations (i.e. using the same communication channels that would be used to teach people). Nerve makes use of existing patterns for networked applications together with the recent Data Distribution Service specification, which is a publish/subscribe standard for real‐time and distributed systems that provides a wide set of QoS policies. In this paper, these different QoS policies have been applied carefully to achieve the best performance of the target robot. Copyright © 2012 John Wiley & Sons, Ltd.