Safe Physical Interaction Research Articles

A coordinated framework of aerial manipulator for safe and compliant physical interaction

The challenge of ensuring compliant behavior in aerial manipulators during physical interactions is addressed in this study. This work presents a coordinated interactive framework for aerial manipulators, specifically designed to ensure compliant physical interactions during contact with the surroundings. Without the reliance on force sensors, an external force estimator is employed to recognize interaction intentions. The coordinated framework leverages a Model Predictive Control (MPC) planner equipped with adaptive weights, facilitating the coordination of movements for both the quadrotor and manipulator. The adaptive adjustment of weights in the cost function allows for the attainment of diverse interaction behaviors in response to unexpected external forces. Notably, the methodology stands out from current control strategies by incorporating autonomous stiffness adjustment in response to interaction forces, thereby optimizing the delicate balance between safety and precision. The passivity of the system is guaranteed by using a Lyapunov-like function during the physical interaction. The effectiveness of the proposed framework is validated by experiments for the scenario of human operator and aerial manipulator collaborative work.

Concept design of a monolithic compliant series-elastic actuator with integrated position and two-level force sensing

Safe physical interaction is significant for the protection of the operated target during delicate micromanipulation tasks. This paper proposes the concept design of a series-elastic actuator mainly constructed by a fully compliant mechanism. It can reduce the transient contact force when the tip touches an object and monitor the interaction force in real time for safe handling. The rear end of the actuator is a flexible linear guide connected with a voice coil motor to offer a bidirectional displacement. The actuator’s front end features a stiffness-switchable mechanism based on folding flexure beams, which provides two-level force buffering and sensing to accommodate varying degrees of contact status. Three strain gauges are integrated into the flexible beam of the mechanism, one of which is used for displacement sensing and the others for monitoring interactive force. After the structural design of the actuator, an analytical model is developed to assess its kinematic performance, which is further verified by finite-element simulation and experimental testing. The experimental results agree with the model and simulation, confirming the effectiveness and reliability of the actuator. The compliant actuator can be applied to some micromanipulation scenarios requiring precision positioning, contact, and interaction at the operating end.

Safe physical interaction with cobots: a multi-modal fusion approach for health monitoring.

Health monitoring is a critical aspect of personalized healthcare, enabling early detection, and intervention for various medical conditions. The emergence of cloud-based robot-assisted systems has opened new possibilities for efficient and remote health monitoring. In this paper, we present a Transformer-based Multi-modal Fusion approach for health monitoring, focusing on the effects of cognitive workload, assessment of cognitive workload in human-machine collaboration, and acceptability in human-machine interactions. Additionally, we investigate biomechanical strain measurement and evaluation, utilizing wearable devices to assess biomechanical risks in working environments. Furthermore, we study muscle fatigue assessment during collaborative tasks and propose methods for improving safe physical interaction with cobots. Our approach integrates multi-modal data, including visual, audio, and sensor- based inputs, enabling a holistic assessment of an individual's health status. The core of our method lies in leveraging the powerful Transformer model, known for its ability to capture complex relationships in sequential data. Through effective fusion and representation learning, our approach extracts meaningful features for accurate health monitoring. Experimental results on diverse datasets demonstrate the superiority of our Transformer-based multi- modal fusion approach, outperforming existing methods in capturing intricate patterns and predicting health conditions. The significance of our research lies in revolutionizing remote health monitoring, providing more accurate, and personalized healthcare services.

Unified Power and Admittance Adaptation for Safe and Effective Physical Interaction With Unmodelled Dynamic Environments

Unified Power and Admittance Adaptation for Safe and Effective Physical Interaction With Unmodelled Dynamic Environments

Touch, press and stroke: a soft capacitive sensor skin

Soft sensors that can discriminate shear and normal force could help provide machines the fine control desirable for safe and effective physical interactions with people. A capacitive sensor is made for this purpose, composed of patterned elastomer and containing both fixed and sliding pillars that allow the sensor to deform and buckle, much like skin itself. The sensor differentiates between simultaneously applied normal force and shear using summation and differences of signals from four deformable capacitors. Cross talk from shear to normal force is less than 2.5%, and between shear axes is less than 10%. Normal and shear stress sensitivity is 0.49 kPa and 0.31 kPa respectively, with a minimum displacement resolution of 40 μm. In addition, finger proximity is detectable at a range of up to 15 mm. The operation is demonstrated on a simple gripper holding a cup. The combination of features and the straightforward fabrication method make this sensor a candidate for implementation as a sensing skin for humanoid robotics applications.

Energy-Efficient Actuator Design Principles for Robotic Leg Prostheses and Exoskeletons: A Review of Series Elasticity and Backdrivability

AbstractRobotic leg prostheses and exoskeletons have traditionally been designed using highly-geared motor-transmission systems that minimally exploit the passive dynamics of human locomotion, resulting in inefficient actuators that require significant energy consumption and thus provide limited battery-powered operation or require large onboard batteries. Here we review two of the leading energy-efficient actuator design principles for legged and wearable robotic systems: series elasticity and backdrivability. As shown by inverse dynamic simulations of walking, there are periods of negative joint mechanical work that can be used to increase efficiency by recycling some of the otherwise dissipated energy using series elastic actuators and/or backdriveable actuators with energy regeneration. Series elastic actuators can improve shock tolerance during foot-ground impacts and reduce the peak power and energy consumption of the electric motor via mechanical energy storage and return. However, actuators with series elasticity tend to have lower output torque, increased mass and architecture complexity due to the added physical spring, and limited force and torque control bandwidth. High torque density motors with low-ratio transmissions, known as quasi-direct drives, can likewise achieve low output impedance and high backdrivability, allowing for safe and compliant human-robot physical interactions, in addition to energy regeneration. However, torque-dense motors tend to have higher Joule heating losses, greater motor mass and inertia, and require specialized motor drivers for real-time control. While each actuator design has advantages and drawbacks, designers should consider the energy-efficiency of robotic leg prostheses and exoskeletons during daily locomotor activities besides continuous level-ground walking.

3-D-Curved Iontronic Tactile Sensor and Denoising Method for Physical Human–Robot Interactions

Tactile information is essential for safe physical interactions between humans and robots. However, for all kinds of tactile sensing principles, it is still challenging to monitor contact status during physical human–robot interactions (pHRIs) on 3-D irregular surfaces due to complex tactile sensor design and signal denoising. This study developed a robotic tactile sensing system by introducing the iontronic electric double-layer capacitor (EDLC) supercapacitive principle, covering a large area ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$191\times191$ </tex-math></inline-formula> mm), and ensuring a dense spatial resolution by using 1024 sensing units. A special grounding strategy and a spatio-temporal (S-T) digital filter significantly reduced the crosstalk interference resulting from low ionic mobility and lateral migration. Additionally, based on surface flattening technology, a 3-D-curved tactile sensor is presented for tactile sensing of a humanoid lower limb’s foot. Finally, response tests based on human actions were performed by the two kinds of robots. Results showed that the system could rapidly capture the entire dynamic process of pHRI actions and distinguish actions in real-time. This study enables the possibility of tactile sensing with whole-body coverage, rapid responses, and tactile-based robotic intelligence.

Development of a low-cost teleoperated and semi-autonomous robotic arm

A safe human-robot physical interaction is required when the robot is used to help humans. This can be achieved by introducing a teleoperated robotic arm in which a human can teach the robot before performing tasks remotely. This paper develops and establishes a three-degree-of-freedom robotic arm and teaching pendant. In particular, a flexible robotic arm is operated in two different modes, namely, the teleoperated mode and the semi-autonomous mode. The teleoperated mode is a manual control using a teaching pendant, where the robot arm replicates various movements of the teaching pendant. On the contrary, the semi-autonomous mode allows the robot to execute a task from one point to another point repetitively after at least one training of the teaching pendant. The Arduino Uno board is employed as a microcontroller, and the integrated development environment ( IDE ) software is used to write and upload the computer code. A series of tests in which the robot performs different tasks is recorded to evaluate the accuracy and consistency of the semi-autonomous and teleoperated modes. The results show that the performance of the proposed low-cost teleoperated robotic arm is reliable and safe to perform various tasks based on the teaching pendant.

Human-Robot Variable Impedance Skills Transfer Learning Based on Dynamic Movement Primitives

Endowing robots with human-like abilities to perform motor skills smoothly and naturally is one of the important goals of robotics. Learning from demonstration (LfD) has been successfully applied for learning tasks on robots, for which the human tutor can demonstrate a successful execution. Learning human stiffness schedule strategy is one of the promising approaches for lightweight collaborative robots to execute physical in-contact tasks while remaining compliant when possible, as it enables robots to show human-like adaptive impedance behavior in an effective and efficient manner. In this letter, we develop a robot skill learning framework considering both movement and impedance features. Electromyography (EMG)-based method is used to estimate human upper limb stiffness such that the information of impedance can be obtained. Dynamic movement primitives (DMPs) model is employed to model movement and impedance features simultaneously. Throughout this procedure, the robot skills learning from a human tutor can be achieved. Besides, the learned skills can be generalized. An impedance controller imitating the human impedance mechanism is utilized in the task reproduction phase to achieve both task successful execution and safe physical interaction. The effectiveness of the proposed methods is verified in a water pumping task on a Kinova robot.

Experimental Evaluation Of Intuitive Programming Of Robot Interaction Behaviour During Kinesthetic Teaching Using sEMG And Cutaneous Feedback

Modern applications related to service or production can nowadays benefit from the introduction of collaborative robots already available on the market and endowed with several advanced features such as precise torque control or safe physical interaction with operators. More importantly, collaborative robots allow operators to teach end-effector trajectories by means of physical interaction – known as kinesthetic teaching – which is one of the most intuitive programming-by-demonstration techniques. However, important functionalities provided by modern collaborative robots, like the possibility of performing smooth interactions, cannot be programmed intuitively with the available framework of kinesthetic teaching. In the present study, we propose and experimentally evaluate a robot programming framework for the simultaneous teaching of both trajectories by means of kinesthetic teaching, and robot interaction behavior by means of impedance shaping along the trajectory exploiting a wearable interface. Specifically, the wearable interface is designed to not affect the free motion of the operator, necessary to perform kinesthetic teaching, and it is based on the usage of surface electromyography (sEMG) and vibrotactile stimulation. In this way, we propose an intuitive robot programming framework for an offline robot trajectory and interaction behavior programming, according to which the operator will be able to plan interactions with the environment and humans. In this article, we report a preliminary experimental evaluation of the proposed system, in which an operator will teach a 7-degrees-of-freedom manipulator the execution of a task on a robotic wiring test-bed. In the experiment, the programming of requested compliance levels during the kinesthetic teaching of a trajectory is performed, and the reported results show that the provided wearable interface is successfully exploited by the operator. Finally, the experiment also demonstrates that the offline intuitive programming of trajectories and impedance levels can be exploited online for human-robot co-work.

Introducing GARMI - A Service Robotics Platform to Support the Elderly at Home: Design Philosophy, System Overview and First Results

This letter introduces GARMI, a service robotics platform conceptualized with a focus on assisting elderly at home. GARMI is designed to provide support with household tasks, as an avatar for tactile-enabled communication and as an interface for telemedicine and emergency assistance. Its unique humanoid design features a sensor-equipped multi-modal head designed for natural human-machine communication as well as a whole-body torque-control interface for safe physical human-machine interaction. GARMI's modular software architecture and distinctive whole-body control scheme allows multimodal dynamic coupling. Additionally, every system component can actively produce as well as sense forces and can thus serve as a haptic feedback interface when interacting with its environment and users. Furthermore, GARMI is the first mobile humanoid designed with specialized use-inspired avatar stations: PARTI for dual-arm-based exoskeleton-like remote-control with force-feedback and MUCKI for transparent remote doctor-patient interaction with both audiovisual and safe haptic feedback. The specialized software architecture allows for rapid prototyping and field-testing of new behaviors for telemedicine, multi-modal interaction and autonomous physical and service assistance. Our first results reveal the potential of our use-driven force-based whole-body control mobile humanoid for daily living and telemedicine scenarios in an elderly care research facility.

Methodology for the definition of the optimal assembly cycle and calculation of the optimized assembly cycle time in human-robot collaborative assembly

Industrial collaborative robotics is an enabling technology and one of the main drivers of Industry 4.0 in industrial assembly. It allows a safe physical and human-machine interaction with the aim of improving flexibility, operator’s work conditions, and process performance at the same time. In this regard, collaborative assembly is one of the most interesting and useful applications of human-robot collaboration. Most of these systems arise from the re-design of existing manual assembly workstations. As a consequence, manufacturing companies need support for an efficient implementation of these systems. This work presents a systematical methodology for the design of human-centered and collaborative assembly systems starting from manual assembly workstations. In particular, it proposes a method for task scheduling identifying the optimal assembly cycle by considering the product and process main features as well as a given task allocation between the human and the robot. The use of the proposed methodology has been tested and validated in an industrial case study related to the assembly of a touch-screen cash register. Results show how the new assembly cycle allows a remarkable time reduction with respect to the manual cycle and a promising value in terms of payback period.

Modeling and control of a 4-ADOF upper-body exoskeleton with mechanically decoupled 3-D compliant arm-supports for improved-pHRI

Safe physical human-robotic interaction is a crucial concern for worn exoskeletons where lower weight requirement limits the number and size of actuators to be used. A novel control strategy is suggested in this paper for the low degree of freedom exoskeletons, by combining proposed mechanically decoupled passive-compliant arm-supports with active compliance, to achieve an improved and safer physical-human-robotic-interaction performance, while considering the practical limitations of low-power actuators. The approach is further improved with a novel vectoral-form of disturbance observer-based dynamic load-torque compensator, proposed to linearize and decouple the nonlinear human-machine dynamics effectively. The design of a four-degree of freedom exoskeleton test-rig that can assure the implementation of the proposed strategy is also shortly presented. It is shown through simulation and experimentation, that the use of proposed strategy results in an improved and safer physical human-robotic interaction, for the exoskeletons using limited-power actuators. It is also shown both through simulation and experimentation, that the proposed vectoral-form of disturbance based dynamic load-toque compensator, effectively outperforms the other traditional compensators in compensating the load-torques at the joints of the exoskeleton.

Design of Human-Centered Collaborative Assembly Workstations for the Improvement of Operators’ Physical Ergonomics and Production Efficiency: A Case Study

Industrial collaborative robotics is one of the main enabling technologies of Industry 4.0. Collaborative robots are innovative cyber-physical systems, which allow safe and efficient physical interactions with operators by combining typical machine strengths with inimitable human skills. One of the main uses of collaborative robots will be the support of humans in the most physically stressful activities through a reduction of work-related biomechanical overload, especially in manual assembly activities. The improvement of operators’ occupational work conditions and the development of human-centered and ergonomic production systems is one of the key points of the ongoing fourth industrial revolution. The factory of the future should focus on the implementation of adaptable, reconfigurable, and sustainable production systems, which consider the human as their core and valuable part. Strengthening actual assembly workstations by integrating smart automation solutions for the enhancement of operators’ occupational health and safety will be one of the main goals of the near future. In this paper, the transformation of a manual workstation for wire harness assembly into a collaborative and human-centered one is presented. The purpose of the work is to present a case study research for the design of a collaborative workstation to improve the operators’ physical ergonomics while keeping or increasing the level of productivity. Results demonstrate that the achieved solution provides valuable benefits for the operators’ working conditions as well as for the production performance of the companies. In particular, the biomechanical overload of the worker has been reduced by 12.0% for the right part and by 28% for the left part in terms of manual handling, and by 50% for the left part and by 57% for the right part in terms of working postures. In addition, a reduction of the cycle time of 12.3% has been achieved.

3D printing of very soft elastomer and sacrificial carbohydrate glass/elastomer structures for robotic applications

Recently, soft materials such as silicone elastomers are widely used in soft robotics due to their high flexibility and safe physical interaction with humans. Focusing on three aspects—highly elastic materials, sacrificial materials, and actuation units, we aim to develop an additive manufacturing strategy that allows the fabrication of highly elastic soft structures more efficiently. First, silicone thinner is used to tailor the mechanical properties of the soft silicone elastomer. We found out that adding 20% volume thinner to the silicone improved the ultimate tensile elongation of 3D printed silicone samples upto 1260%, which is the highest among all the 3D printed soft materials previously reported. However, in a cyclic tensile test, this strain is not achieved; instead, the maximum strain was 600%. Second, carbohydrate glasses are introduced as sacrificial materials for 3D printing silicone with hollow channels. Few configurations of pneumatic and hydraulic actuators are developed by forming channels in the silicone elastomer via 3D printed sacrificial carbohydrate structures and subsequently dissolving. The 3D printed structures actuated successfully to get morphed shapes, which showed that our method is effective for manufacturing of functional soft robotic structures.

Design and development of a soft robotic manipulator

Soft robotic manipulator, which has advantages of low-weight, flexibility, safe physical interaction, shows a wide prospect in both industry and family application. In this paper, a design of pneumatic soft manipulator, in which the modularization concept is adopted, is presented. The soft manipulator is made up of three pneumatic soft actuators stacked in series. The soft actuator is composed of three inflatable bellows, which are connected and constrained by a series of thin rigid plates. The bellows is stretched by the actuation of pressured gas. The cooperation of the bellows and the rigid constraints effect the soft actuator’s directional move. The stacking connection of several soft actuators can have a large coverage in space. Based on this design, the kinematic models of the soft actuator as well as the soft manipulator are proposed using piecewise constant curvature method. The elastic coefficient of the actuator, which is necessary in the numerical calculation of the proposed model, is determined experimentally. Finally, the prototype and experimental set up of the soft manipulator are designed and developed. The correctness of the proposed model and the end-repeat positioning accuracy are validated experimentally.

A Variational Approach to Minimum-Jerk Trajectories for Psychological Safety in Collaborative Assembly Stations

Collaborative robotics is one of the main drivers of the fourth industrial revolution, where robots and humans cooperate in a shared workspace. Such robots are distinguished from the traditional robots by several features that, if tuned correctly, ensure safe physical interaction. However, interaction with robots also represents a potential source of psychological stress. Indeed, it has been proven that working and sharing the workspace with robots can psychologically stress the operator. This drawback—in addition to the reduced speed operation imposed by safety standards—represents a constraint difficult to bear for small and medium-sized enterprises. In this letter, we propose a novel trajectory-planning methodology for collaborative robots able to ensure both psychological and physical safety. We rely on the available literature and the calculus of variations to identify a family of psychologically acceptable minimum-jerk trajectories. From the identified family, we compute the fastest trajectory that limits the physical danger represented by the robot. A numerical implementation of the method is discussed, and experimental results prove the effectiveness of the idea.

Using EEG for Predicting User Preferences of Physical Compliance in Human-Robot Cooperation

Using EEG for Predicting User Preferences of Physical Compliance in Human-Robot Cooperation

Highly dexterous 2-module soft robot for intra-organ navigation in minimally invasive surgery.

For some surgical interventions, like the Total Mesorectal Excision (TME), traditional laparoscopes lack the flexibility to safely maneuver and reach difficult surgical targets. This paper answers this need through designing, fabricating and modelling a highly dexterous 2-module soft robot for minimally invasive surgery (MIS). A soft robotic approach is proposed that uses flexible fluidic actuators (FFAs) allowing highly dexterous and inherently safe navigation. Dexterity is provided by an optimized design of fluid chambers within the robot modules. Safe physical interaction is ensured by fabricating the entire structure by soft and compliant elastomers, resulting in a squeezable 2-module robot. An inner free lumen/chamber along the central axis serves as a guide of flexible endoscopic tools. A constant curvature based inverse kinematics model is also proposed, providing insight into the robot capabilities. Experimental tests in a surgical scenario using a cadaver model are reported, demonstrating the robot advantages over standard systems in a realistic MIS environment. Simulations and experiments show the efficacy of the proposed soft robot.

Whole-body multi-contact motion in humans and humanoids: Advances of the CoDyCo European project

Traditional industrial applications involve robots with limited mobility. Consequently, interaction (e.g. manipulation) was treated separately from whole-body posture (e.g. balancing), assuming the robot firmly connected to the ground. Foreseen applications involve robots with augmented autonomy and physical mobility. Within this novel context, physical interaction influences stability and balance. To allow robots to surpass barriers between interaction and posture control, forthcoming robotic research needs to investigate the principles governing whole-body motion and coordination with contact dynamics. There is a need to investigate the principles of motion and coordination of physical interaction, including the aspects related to unpredictability. Recent developments in compliant actuation and touch sensing allow safe and robust physical interaction from unexpected contact including humans. The next advancement for cognitive robots, however, is the ability not only to cope with unpredictable contact, but also to exploit predictable contact in ways that will assist in goal achievement. Last but not least, theoretical results needs to be validated in real-world scenarios with humanoid robots engaged in whole-body goal-directed tasks. Robots should be capable of exploiting rigid supportive contacts, learning to compensate for compliant contacts, and utilizing assistive physical interaction from humans. The work presented in this paper presents state-of-the-art in these domains as well as some recent advances made within the framework of the CoDyCo European project.