Human Collaboration Research Articles

Contribution of Human Augmentation in Creating Cognitive and Physical Improvements as an Integral Part of the Human Body

This article is based on the contribution of human augmentation in creating physical and cognitive improvement. Nowadays it makes a positive impact on the medical sector and technological advancement provides outstanding opportunities to lead a better life. It helps to increase the action of the people and also increase the level of activity. Artificial intelligence helps to improve the facilities of technologies and there are various kinds of sensors, devices, and another sensory mechanism that develop the action of the technologies. This study has focused on the construct between human enhancement and human augmentation. This study has also shed light on several kinds of human augmentation and also focused on the collaboration of humans and machines. It has been discussed that neuromorphic technologies achieved so much attention from scientists and it can be considered as a computing system that can mimic the “neurobiological architecture”. This study has focused on several kinds of several kinds of technologies that can improve the individuals’ difficulties such as artificial limbs, bionics, and other. On the other hand, this study has also elaborated the bonding between machine and collaboration. The “human augmentation” is the combination of biological applications and human beings in the field of medical science. This kind of advancement incorporates with medicines, gene editing, wearable devices and cellular implants. These technological experiments can be implemented on the body of human being and researchers have been working on the area of smart application of drugs to prohibit some serious health complications. These experiments can improve the society as well as make an impact on the people psychology as well. Keywords : Artificial limbs, genetic alteration, HUMAN AUGMENTATION, human-machine collaboration, implementation of chips, neurobiological architecture plastic surgery, Prosthetics.

A fusion-based spiking neural network approach for predicting collaboration request in human-robot collaboration

• Compared with the traditional assembly method, a fusion-based spiking neural network approach for predicting collaboration request proposed in this paper has the following advantages. • A human collaboration request prediction method under the joint influence of multi-channel information is proposed. • The FSNN method for asynchronous monitoring is proposed to reduce the workload of non-time-series feature detection. • The coding methods of different input information, and the threshold adaptive updating method under the influence of different neurons are proposed. • Compared with different prediction methods, the superiority of the FSNN scheme is proved. In human-robot collaborative (HRC) manufacturing systems, how the collaborative robots engage in the collaborative tasks and complete the corresponding work in a timely manner according to the actual state has been a critical factor that hinders the efficiency of HRC. Inappropriate collaborative behaviors will result in a poor perceptual experience for human operators (e.g., robots starting action too early or too late). To address this issue, a fusion-based spiking neural networks (FSNNs) approach for collaboration request prediction is proposed, aiming to find the right collaboration timing for robots in HRC assembly system and to minimize human aversion without affecting human operation behaviors. By encoding human behavior, product state and robot pose into spiking signals that can be processed by FSNNs, the spatio-temporal coupling relationship between those three aspects can be comprehensively analyzed, and to solve the appropriate timing of robot participation in collaboration. Finally, demonstrative experiments are carried out on the HRC assembly of generator end caps in the lab environment. Compared with the baseline methods, the decision accuracy of the proposed one is improved by nearly 30%, which further proves its effectiveness.

Sexy, slimy, monstrous: Infection as collaboration in Ben Wheatley’s In the Earth and Jaco Bouwer’s Gaia

This article offers a discussion of two eco-horror feature films, each released in 2021: Ben Wheatley’s In the Earth, and Jaco Bouwer’s Gaia. With a pandemic theme, both film fictions feature sentient woods and forests which infect humans with fungal spores. This is the non-human world enlisting human collaboration. Both films break the mould of the traditional eco-horror; invariably, the latter dramatizes how the natural world avenges itself on humankind. This storytelling mode involves the use of catastrophe narratives. Elizabeth Parker has explored the alternative genre, the ‘ecoGothic’ (), which stimulates ambiguous emotional responses to human–non-human hybrid forms. To date, there is little scholarship which analyses such hybrids through the lens of queer ecology. Therefore, I will bring Timothy Morton’s theory of queer ecology (), dark ecology () and his concept of the spectral non-human world () into dialogue with Simon C. Estok’s term ‘the slimic imagination’ (). My thesis is that both films engage an interaction between queer and slimic narratives in order to undo the catastrophe narrative. Queer slime is that which destabilizes gender binaries as a means of creating uneasy but intimate collaborations between the human and non-human world.

Goal-Conditioned Reinforcement Learning within a Human-Robot Disassembly Environment

The introduction of collaborative robots in industrial environments reinforces the need to provide these robots with better cognition to accomplish their tasks while fostering worker safety without entering into safety shutdowns that reduce workflow and production times. This paper presents a novel strategy that combines the execution of contact-rich tasks, namely disassembly, with real-time collision avoidance through machine learning for safe human-robot interaction. Specifically, a goal-conditioned reinforcement learning approach is proposed, in which the removal direction of a peg, of varying friction, tolerance, and orientation, is subject to the location of a human collaborator with respect to a 7-degree-of-freedom manipulator at each time step. For this purpose, the suitability of three state-of-the-art actor-critic algorithms is evaluated, and results from simulation and real-world experiments are presented. In reality, the policy’s deployment is achieved through a new scalable multi-control framework that allows a direct transfer of the control policy to the robot and reduces response times. The results show the effectiveness, generalization, and transferability of the proposed approach with two collaborative robots against static and dynamic obstacles, leveraging the set of available solutions in non-monotonic tasks to avoid a potential collision with the human worker.

Leader–Follower Dynamics in Complex Obstacle Avoidance Task

A question that many researchers in social robotics are addressing is how to create more human-like behaviour in robots to make the collaboration between a human and a robot more intuitive to the human partner. In order to develop a human-like collaborative robotic system, however, human collaboration must first be better understood. Human collaboration is something we are all familiar with, however not that much is known about it from a kinematic standpoint. One dynamic that hasn’t been researched thoroughly, yet naturally occurs in human collaboration, is for instance leader–follower dynamics. In our previous study, we tackled the question of leader–follower role allocation in human dyads during a collaborative reaching task, where the results implied that the subjects who performed higher in the individual experiment would naturally assume the role of a leader when in physical collaboration. In this study, we build upon the leader–follower role allocation study in human dyads by observing how the leader–follower dynamics change when the collaborative task becomes more complex. Here, the study was performed on a reaching task, where one subject in a dyad was faced with an additional task of obstacle avoidance when performing a 2D reaching task, while their partner was not aware of the obstacle. We have found that subjects change their roles throughout the task in order to complete it successfully, however looking at the overall task leader the higher-performing individual will always dominate over the lower-performing one, regardless of whether they are aware of the additional task of obstacle avoidance or not.

Asymmetric Cooperation Control of Dual-Arm Exoskeletons Using Human Collaborative Manipulation Models.

The exoskeleton is mainly used by subjects who suffer muscle injury to enhance motor ability in the daily life environment. Previous research seldom considers extending human collaboration skills to human-robot collaborations. In this article, two models, that is: 1) the following the better model and 2) the interpersonal goal integration model, are designed to facilitate the human-human collaborative manipulation in tracking a moving target. Integrated with dual-arm exoskeletons, these two models can enable the robot to successfully perform target tracking with two human partners. Specifically, the manipulation workspace of the human-exoskeleton system is divided into a human region and a robot region. In the human region, the human acts as the leader during cooperation, while, in the robot region, the robot takes the leading role. A novel region-based Barrier Lyapunov function (BLF) is then designed to handle the change of leader roles between the human and the robot and ensures the operation within the constrained human and robot regions when driving the dual-arm exoskeleton to track the moving target. The designed adaptive controller ensures the convergence of tracking errors in the presence of region switches. Experiments are performed on the dual-arm robotic exoskeleton for the subject with muscle damage or some degree of motor dysfunctions to evaluate the proposed controller in tracking a moving target, and the experimental results demonstrate the effectiveness of the developed control.

Real-time traffic impedance and priority based cooperative path planning mechanism for SOC-ITS: Efficiency and equilibrium

With the support of new information technologies such as the Internet of Vehicles (IoV) and cloud computing, the Intelligent Transportation System (ITS) has begun a new stage role as Cooperative ITS (C-ITS). Moreover, featuring the collaboration of vehicles, roads, clouds and humans, it becomes possible to scientifically mitigate several serious phenomena in the traditional ITS, including traffic congestion, blocked emergency vehicles, etc. While focusing on optimizing the Quality of Transportation Service (QoTS) requirements from vehicles with different types of tasks, covering emergency vehicles (fire engine, police vehicle and ambulance), shuttle buses, and others, this paper studies and proposes a new service-oriented and cloud-based cooperative path planning mechanism for such heterogeneous vehicles. First, the architecture of service-oriented cooperative path planning is established, where the types and QoTS of traffic tasks are illustrated. Then, a hybrid traffic impedance model is put forward, in which three influencing factors of traffic impedance are concluded and combined through the entropy weight method. On this basis, a service-oriented cooperative path planning mechanism with load balancing strategies named SoCPPLB is designed. The main principle of SoCPPLB is to provide different levels of path planning service for vehicles according to their QoTS requirements per the constraint of the QoTS of vehicles and traffic equilibrium. Finally, all these proposed models and algorithms have been implemented within the Simulation of Urban MObility (SUMO), and further verified under several typical traffic scenarios. The experimental results show that, this new mechanism can effectively improve the QoTS of emergency vehicles and traffic equilibrium performance obviously.

Teaching agents to understand teamwork: Evaluating and predicting collective intelligence as a latent variable via Hidden Markov Models

Rapid growth in the reliance on teamwork in organizations, coupled with advances in artificial intelligence, has fueled increased use of Human Autonomy Teams (HATs) involving the collaboration of humans and agents to complete work. Although there are many successful examples of HATs, researchers and technology developers can see additional applications if agents were better able to understand the mental states of humans to anticipate what a team is likely to do next. Creating this capability requires the creation of models of team interaction that enable agents to interpret a team’s current state and anticipate its future state. To build this model, we draw on research on collective intelligence (CI), which shows a team’s capability to work together can be characterized by a latent collective intelligence factor, based on observations of work across a range of tasks, and which predicts a team’s ability to accomplish a wide range of goals in the future. While some work uses a specific battery of CI tasks, more recent studies have identified observable collaborative process metrics that can be captured passively. Building on this work, we propose a method of evaluating CI by representing it as a latent variable represented by the hidden state in a Hidden Markov Model. The observations used as input to the model are the team’s observable collaborative process behaviors (i.e., collective effort, use of task-related skills, and task-strategy efficiency). We show by learning the set of hidden states representing a team’s observed collaborative process behaviors over time, we both learn information about the team’s CI, predict how CI will evolve in the future, and suggest when an agent might intervene to improve team performance. Based on the model’s observations, we discuss how it can help agents diagnose teamwork and possibly make interventions to improve CI by identifying areas of collaborative process (collective effort, skill use, or task strategy) that could be improved.

Measuring Interaction Bandwidth During Physical Human-Robot Collaboration

An assistive robot can augment human performance by providing physical assistance or motion guidance. In human-robot collaboration, it is important to know the bandwidth of an individual's ability to generate motion in response to external stimuli, such as visual or haptic cuing. This becomes particularly relevant in timing-sensitive tasks, such as walking or catching a falling object. In this work, we propose a frequency-based assessment of motion that enables us to measure the bandwidth of physical human-robot interaction (pHRI)—quantifying how fast individuals can respond to stimuli on a continuous basis. We introduce a robot-assisted virtual dynamic task with a tunable resonant frequency. A human subject study with seven participants shows that our task can elicit a dynamic response in a participant at frequencies of 0.5 Hz, 1 Hz, 1.5 Hz and 2.5 Hz at the arm. Using the virtual task, we test whether haptic cues improve motion timing. At all tested frequencies, we find that haptic stimuli help guide timing of dynamic movement and improve performance compared to visual-only cuing. By quantifying the interaction bandwidth for other pHRI systems—particularly when the human collaborators have neuromotor impairments—our method can help assistive robots adapt to an individual. Moreover, our results highlight the importance of incorporating haptic feedback into pHRI for dynamic tasks—haptics can provide guidance around motion timing, such as in assistive robots used for assessment and physical rehabilitation.

Parametric PH Curves Model Based Kinematic Control of the Shape of Mobile Soft Manipulators in Unstructured Environment

Mobile soft continuum manipulator (MSCM) is more and more used in various applications of everyday life, such as logistics, farming, medical diagnosis, medical therapy, bakery, human collaboration, etc. However, the control of its shape remains a major challenge, especially when it is navigating in an unstructured environment. In this contribution, a kinematic-model-based shape control is designed in the space configuration for an MSCM, allowing its autonomous navigation in the presence of obstacles. For that purpose, the shape kinematic modeling is realized using a parametric spatial Pythagorean hodograph (PH) curve with a predefined length, whereas the artificial potential field algorithm acts on the control points of the PH curve for shape adaptation in presence of obstacles. The novelty of the proposed work resides in the use of parametric curve formalism to model the shape of an infinite degree of freedom mobile soft manipulator robot and its kinematic control by optimizing its bending potential energy during its motion. The shape optimization based on a minimal bending energy during the collision-free path is respecting a sliding mode strategy applied to the PH curve control points. Experimental results are obtained using a class of MSCM called RobotinoXT. These experiments show the advantages of using a reduced kinematic model based on PH curve to control the MSCM shape in dynamic motion.

Collaborating eye to eye: Effects of workplace design on the perception of dominance of collaboration robots.

The concept of Human-Robot Collaboration (HRC) describes innovative industrial work procedures, in which human staff works in close vicinity with robots on a shared task. Current HRC scenarios often deploy hand-guided robots or remote controls operated by the human collaboration partner. As HRC envisions active collaboration between both parties, ongoing research efforts aim to enhance the capabilities of industrial robots not only in the technical dimension but also in the robot’s socio-interactive features. Apart from enabling the robot to autonomously complete the respective shared task in conjunction with a human partner, one essential aspect lifted from the group collaboration among humans is the communication between both entities. State-of-the-art research has identified communication as a significant contributor to successful collaboration between humans and industrial robots. Non-verbal gestures have been shown to be contributing aspect in conveying the respective state of the robot during the collaboration procedure. Research indicates that, depending on the viewing perspective, the usage of non-verbal gestures in humans can impact the interpersonal attribution of certain characteristics. Applied to collaborative robots such as the Yumi IRB 14000, which is equipped with two arms, specifically to mimic human actions, the perception of the robots’ non-verbal behavior can affect the collaboration. Most important in this context are dominance emitting gestures by the robot that can reinforce negative attitudes towards robots, thus hampering the users’ willingness and effectiveness to collaborate with the robot. By using a 3 × 3 within-subjects design online study, we investigated the effect of dominance gestures (Akimbo, crossing arms, and large arm spread) working in a standing position with an average male height, working in a standing position with an average female height, and working in a seated position on the perception of dominance of the robot. Overall 115 participants (58 female and 57 male) with an average age of 23 years evaluated nine videos of the robot. Results indicated that all presented gestures affect a person’s perception of the robot in regards to its perceived characteristics and willingness to cooperate with the robot. The data also showed participants’ increased attribution of dominance based on the presented viewing perspective.

Three Stage Drawing Transfer

This project creates a visual-mental-physical circuit between a Generative Adversarial Network (GAN), a co-robotic arm, and a five-year-old child. From training images to the latent space of a GAN, through pen on paper to a live human collaborator, it establishes a series of translational stages between humans and non-humans played out through the medium of drawing. Trained on a subset of the Rhoda Kellogg Child Art Collection, the neural network at the center of this piece learns its own representations of these images. The generated results, synthetic children's drawings, are of interest both for being outside of adult conventions and learned expression-like Dubuffet's art brut or Surrealist automatism-and for how they align machine learning with the human act of learning to draw. The project layers many kinds of agency and embodiment: from the thousands of anonymous children who produced the original artwork used as training data, through the co-robot drawing from GAN-generated imagery, to the human child's active perception and graphic response to the robot. These questions of where we search for the other; when we attribute autonomy and intelligence; and why we might wish to escape our human subjectivities speak to core issues in the design and use of AI systems. This project is one attempt to think through those questions in an embodied way.

Improving the detection of community smells through socio‐technical and sentiment analysis

AbstractOpen source software development is regarded as a collaborative activity in which developers interact to build a software product. Such a human collaboration is described as an organized effort of the “social” activity of organizations, individuals, and stakeholders, which can affect the development community and the open source project health. Negative effects of the development community manifest typically in the form of community smells, which represent symptoms of organizational and social issues within the open source software development community that often lead to additional project costs and reduced software quality. Recognizing the advantages of the early detection of potential community smells in a software project, we introduce a novel approach that learns from various community organizational, social, and emotional aspects to provide an automated support for detecting community smells. In particular, our approach learns from a set of interleaving organizational–social and emotional symptoms that characterize the existence of community smell instances in a software project. We build a multi‐label learning model to detect 10 common types of community smells. We use the ensemble classifier chain (ECC) model that transforms multi‐label problems into several single‐label problems, which are solved using genetic programming (GP) to find the optimal detection rules for each smell type. To evaluate the performance of our approach, we conducted an empirical study on a benchmark of 143 open source projects. The statistical tests of our results show that our approach can detect community smells with an average F‐measure of 93%, achieving a better performance compared to different state‐of‐the‐art techniques. Furthermore, we investigate the most influential community‐related metrics to identify each community smell type.

Sustainable Human–Machine Collaborations in Digital Transformation Technologies Adoption: A Comparative Case Study of Japan and Germany

The Digital Transformation (DX) megatrend is fundamentally disrupting and changing the nature of work, business, and industry at a rapid pace. Although the notion of DX has garnered much research interest from practitioners, scholarship on this topic is somehow lagging behind, possibly because of the lack of theoretical frameworks on DX. Recently, most Japanese firms have begun to use diverse digital technologies to sustain their competitive advantages. However, the return of investment on digital technologies has not been as high as expected for some firms. Furthermore, as the visions of Industry 5.0 describe sustainable, resilient, and human-centered future factories that will require smart and resilient capabilities both from next-generation manufacturing systems and human operators, it is necessary to design resilient human–machine collaborations within factories. To this end, this paper presents a research model between DX technologies and scientific problem-solving in terms of deduction, induction, and abduction inference structures as an approach to resilient human–machine collaborations. The purpose of this research is to analyze the difference in the utilization pattern of the digital technology of American, German, and Japanese firms based on three types of decision-making methods. Next, we apply this framework in a comparative case study of two Japanese firms and one German firm, where we find that there is a difference in DX technologies utilization among the Japanese and German firms. We assert that the utilization of IoT technology in the United States and Germany is pursuing IoT with the aim of autonomous control, whereas Japanese firms prioritize robot–human collaboration. Finally, we discuss how our findings contribute to the burgeoning field of resilient human–machine collaborations by showing the distinct roles of deduction, induction, and abduction inference structures. Furthermore, our research contributes to international comparative studies to identify the difference in national IT utilization. Lessons and implications are discussed.

Training and Development in Times of Automation: A Conceptual Roadmap for HRM

As automation of machines proceeds with high acceleration, human resource management (HRM) is challenged to prepare the organization for a “real” collaboration of humans and machines. In order to develop appropriate coping strategies and control systems for HRM, we design a conceptual roadmap for training and development in times of automation in the near and far future. We equally draw from management and computer science research by combining technical-centered and human-centered approaches for training and development. By conceiving humans and machines as complementary actors in the organizations of the future, we suggest four training directions. Our roadmap emphasizes that a fully integrated management of human and machine resources enables HRM to optimize and evaluate the collaborative development of machines and humans. Our research underlines the importance of combining human resource training and development with the progression of automation.

A semantic-centered cloud control framework for autonomous unmanned system

Rich semantic information in natural language increases team efficiency in human collaboration, reduces dependence on high precision data information, and improves adaptability to dynamic environment. We propose a semantic centered cloud control framework for cooperative multi-unmanned ground vehicle (UGV) system. Firstly, semantic modeling of task and environment is implemented by ontology to build a unified conceptual architecture, and secondly, a scene semantic information extraction method combining deep learning and semantic web rule language (SWRL) rules is used to realize the scene understanding and task-level cloud task cooperation. Finally, simulation results show that the framework is a feasible way to enable autonomous unmanned systems to conduct cooperative tasks.

360o 라이다를 이용한 CNN 기반 사람 인식 및 확장 칼만 필터 기반 위치 추적

The collaboration of robots and humans sharing workspace, can increase productivity and reduce production costs. However, occupational accidents resulting in injuries can increase, by removing the physical safety around the robot, and allowing the human to enter the workspace of the robot. In preventing occupational accidents, studies on recognizing humans, by installing various sensors around the robot and responding to humans, have been proposed. Using the LiDAR (Light Detection and Ranging) sensor, a wider range can be measured simultaneously, which has advantages in that the LiDAR sensor is less impacted by the brightness of light, and so on. This paper proposes a simple and fast method to recognize humans, and estimate the path of humans using a single stationary 360° LiDAR sensor. The moving object is extracted from background using the occupied grid map method, from the data measured by the sensor. From the extracted data, a human recognition model is created using CNN machine learning method, and the hyper-parameters of the model are set, using a grid search method to increase accuracy. The path of recognized human is estimated and tracked by the extended Kalman filter.

When AI moderates online content: effects of human collaboration and interactive transparency on user trust

Abstract Given the scale of user-generated content online, the use of artificial intelligence (AI) to flag problematic posts is inevitable, but users do not trust such automated moderation of content. We explore if (a) involving human moderators in the curation process and (b) affording “interactive transparency,” wherein users participate in curation, can promote appropriate reliance on AI. We test this through a 3 (Source: AI, Human, Both) × 3 (Transparency: No Transparency, Transparency-Only, Interactive Transparency) × 2 (Classification Decision: Flagged, Not Flagged) between-subjects online experiment (N = 676) involving classification of hate speech and suicidal ideation. We discovered that users trust AI for the moderation of content just as much as humans, but it depends on the heuristic that is triggered when they are told AI is the source of moderation. We also found that allowing users to provide feedback to the algorithm enhances trust by increasing user agency.

Cystic Echinococcosis: An Impact Assessment of Prevention Programs in Endemic Developing Countries in Africa, Central Asia, and South America

Background. Cystic echinococcosis (CE), caused by the tapeworm species, Echinococcus granulosus sensu stricto (G1), is one of many primary neglected zoonoses worldwide. Within endemic developing countries, CE has multiple effects on animal and human health and well-being. To address such effects, veterinary and human medical sector collaboration on prevention program delivery is essential. To begin preliminary evaluations of county specific prevention programs, a critically appraised topic (CAT) was conducted. It sought to answer: What impact do CE prevention programs have on human and animal disease prevalence, in populations living in endemic developing countries within Africa, Central Asia, and South America? Methodology. The aim was to assess the ability of prevention and control program outputs to produce measurable differences in health, social, and economic outcomes (e.g., improved access to medical services, positive behavioral change, or reduced treatment costs, respectively). Included articles were obtained using predefined inclusion/exclusion criteria from the four databases (CAB Abstracts and Global Health; the National Library of Medicine (PubMed); ScienceDirect; and WHO Institutional Repository of Information Sharing (IRIS)). The articles were appraised using three checklists: the Royal College of Veterinary Surgeons (RCVS), the Critical Appraisals Skills Programme (CASP), and the Joanna Briggs Institute checklists. Results. Ten articles were selected. Geographically, 20% of studies were conducted in South America, 30% in Africa, and 50% in Central Asia. For definitive hosts, dogs, CoproELISA antigen testing, before and after Praziquantel (PZQ) de-worming, was a primary focus. For humans, who are intermediate hosts (IH), disease surveillance methods, namely ultrasound (US), were commonly assessed. Whilst for sheep, also acting as IH, disease prevention methods, such as the EG95 livestock vaccine and de-worming farm dogs, were evaluated. Common to all studies were issues of program sustainability, in terms of regular human US screening, dog de-worming, and annual sheep vaccination. This was attributed to transient and remote human or animal populations; limited access to adequate roads or hospitals; few skilled health workers or veterinarians; an over-reliance on communities to administer preventatives; and limited resources. Conclusion. Despite variations in result validity and collection periods, useful comparisons of CE endemic countries produced key research and program recommendations. Future research recommendations included testing the significance of multiple program outcomes in relation to prevalence (e.g., the social outcome: behavioral change), further research on the impact of livestock vaccinations, and the CE transmission role of waterways and sanitation. Program recommendations included calculating and distinguishing between stray versus owned dog populations; formal representation of internal and external stakeholder interests through institutional organization; establishing sustainable guidelines around the frequency of PZQ and vaccination administration; improved veterinary-human medical training and resource sharing; and combined prevention methods and multiple canine disease management.

An electromyography signals-based human-robot collaboration method for human skill learning and imitation

Human beings have strong perception and decision-making ability. In human-human collaboration, the follower can adjust the motion speed by perceiving the change of external force information, and cooperate with the leader to complete the collaboration task. Transferring human collaboration skills to robots will help improve the flexibility and compliance of human-robot collaboration. For this purpose, an Electromyography (EMG) signals-based human-robot collaboration method is proposed. In the collaboration method, the human arm force, obtained by EMG signals and joint angles, is taken as the interface of human-robot interaction, and the robot is controlled to complete the collaboration task by learning the speed adjustment skills of human tutors. To reduce the inaccuracy and instability of human-robot interaction caused by the fluctuation of EMG signals, the adaptive data correction unit based on tremor information and the input-output control unit based on Naive Bayes are added to the Long Short-Term Memory (LSTM) neural network, and the parallel network structure is used to estimate the three-dimensional arm force. Aiming at the problem that the traditional collaborative control model cannot balance the accuracy and rapidity, a multi-model Gaussian process regression algorithm is used to capture the collaboration skills from inaccurate human-human demonstrations in the way of probability estimation. Finally, taking peg-in-hole assembly as an example, the proposed human-robot collaboration method is verified. Compared with the traditional adaptive admittance control, the proposed collaboration method shows better interaction performance, and its assembly time and assembly success rate are improved.