Comparison of Point Cloud Registration Techniques on Scanned Physical Objects.

This paper presents a comparative analysis of six prominent registration techniques for solving CAD model alignment problems. Unlike the typical approach of assessing registration algorithms with synthetic datasets, our study utilizes point clouds generated from the Cranfield benchmark. Point clouds are sampled from existing CAD models and 3D scans of physical objects, introducing real-world complexities such as noise and outliers. The acquired point cloud scans, including ground-truth transformations, are made publicly available. This dataset includes several cleaned-up scans of nine 3D-printed objects. Our main contribution lies in assessing the performance of three classical (GO-ICP, RANSAC, FGR) and three learning-based (PointNetLK, RPMNet, ROPNet) methods on real-world scans, using a wide range of metrics. These include recall, accuracy and computation time. Our comparison shows a high accuracy for GO-ICP, as well as PointNetLK, RANSAC and RPMNet combined with ICP refinement. However, apart from GO-ICP, all methods show a significant number of failure cases when applied to scans containing more noise or requiring larger transformations. FGR and RANSAC are among the quickest methods, while GO-ICP takes several seconds to solve. Finally, while learning-based methods demonstrate good performance and low computation times, they have difficulties in training and generalizing. Our results can aid novice researchers in the field in selecting a suitable registration method for their application, based on quantitative metrics. Furthermore, our code can be used by others to evaluate novel methods.

A UWB-Ego-Motion Particle Filter for Indoor Pose Estimation of a Ground Robot Using a Moving Horizon Hypothesis.

Ultra-wideband (UWB) has gained increasing interest for providing real-time positioning to robots in GPS-denied environments. For a robot to act on this information, it also requires its heading. This is, however, not provided by UWB. To overcome this, either multiple tags are used to create a local reference frame connected to the robot or a single tag is combined with ego-motion estimation from odometry or Inertial Measurement Unit (IMU) measurements. Both odometry and the IMU suffer from drift, and it is common to use a magnetometer to correct the drift on the heading; however, magnetometers tend to become unreliable in typical GPS-denied environments. To overcome this, a lightweight particle filter was designed to run in real time. The particle filter corrects the ego-motion heading and location drift using the UWB measurements over a moving horizon time frame. The algorithm was evaluated offline using data sets collected from a ground robot that contains line-of-sight (LOS) and non-line-of-sight conditions. An RMSE of 13 cm and 0.12 (rad) was achieved with four anchors in the LOS condition. It is also shown that it can be used to provide the robot with real-time position and heading information for the robot to act on it in LOS conditions, and it is shown to be robust in both experimental conditions.

The impact of an active and passive industrial back exoskeleton on functional performance.

Due to differences in actuation and design, active and passive industrial back exoskeletons could influence functional performance, i.e., work performance, perceived task difficulty, and discomfort, differently. Therefore, this study investigated and compared the impact of the active CrayX (7 kg) and passive Paexo Back (4.5 kg) on functional performance. Eighteen participants performed twelve work-related tasks with both types of exoskeletons and without (NoExo). The CrayX hindered work performance up to 22% in multiple tasks, compared to the Paexo Back and NoExo, while work performance between NoExo and the Paexo Back condition was more comparable, except for stair climbing (13% hindrance). Perceived task difficulty and discomfort seldomly varied between both exoskeletons. Although the CrayX shows promise to benefit workers, limitations in hindrance and comfort should first be addressed. The Paexo Back has demonstrated an advantage in certain static tasks. However, increasing its potential across a broader range of tasks seems warranted.Practitioner Summary: Differences between industrial back exoskeletons with regard to functional performance, i.e. work performance, discomfort and perceived task difficulty, were investigated by evaluating the active CrayX and passive Paexo Back back exoskeletons. The CrayX significantly hindered functional performance, while the Paexo Back seldomly affected functional performance.Abbreviations: WMSD: Work-related musculoskeletal disorder; NoExo: No Exoskeleton; GD: General discomfort; PTD: Perceived task difficulty; BMI: Body Mass Index.

Robotically Aided Method to Characterise the Soft Tissue Interaction with Wearable Robots.

Wearable robots are widely used to enhance, support or assist humans in different tasks. To accomplish this scope, the interaction between the human body and the device should be comfortable, smooth, high-efficient to transfer forces, and safe for the user. Nevertheless, the pressure and shear stress related to these goals have been overlooked or partially analysed. In this sense, it is crucial to understand the soft tissue response through the in-vivo characterisation of multiple areas of the human body. In fact, soft tissue characterisation plays an essential role in calculating the pressure distribution and shear stress. However, current approaches to estimating soft tissue properties are unsuitable for deployment with multiple human body areas. Hence, this work presents a novel methodology to ease the characterisation of soft tissues using a robotic arm and a 3D superficial scanner. First, the robotic arm is validated by comparing the tensile and compression tests to the indentation tests done by the robot, estimating a 10,4% error. The preliminary experimental tests present the hyperelastic model which fit two adjacent zones of the forearm. This analysis can be extended in several ways, such as: calculating the shear stress, the energy losses or deformations caused by the interaction, and investigating the pressure distribution of different types of physical interfaces.

The Dynamic Adoption Journey: A Typology for Users and Non-Users of Occupational Exoskeletons.

Various barriers prevent the adoption of occupational exoskeletons. It is therefore important to understand why some people are willing to use occupational exoskeletons, while others are not. To identify why people use or do not use exoskeletons, we created a typology describing different types of users and non-users. These types were created based on existing literature on internet adoption and social robots. Next, literature and empirical data were used to identify reasons why some people use exoskeletons and others do not use them (yet). The typology includes users with pain and users without work-related musculoskeletal disorders, but also non-users: resisters, rejecters, discontinuers, excluded or expelled non-users. It can be used by companies interested in implementing exoskeletons to identify possible early adopters. For exoskeleton designers, it can be used as a tool to identify non-users and focus on design strategies to enable non-users to become users (such as making exoskeletons that would fit people with a wide range of body shapes). Future research can use these types to identify users and non-users in field trials or pilot projects.

Antropo: An open-source platform to increase the anthropomorphism of the Franka Emika collaborative robot arm.

Robot-to-human communication is important for mutual understanding during human-robot collaboration. Most of the current collaborative robots (cobots) are designed with low levels of anthropomorphism. Therefore, the ability of cobots to express human-like communication is limited. In this work, we present an open-source platform named Antropo to increase the level of anthropomorphism of Franka Emika-a widely used collaborative robot arm. The Antropo platform includes three modules: a camera module for expressing eye gaze, a light module for visual feedback, and a sound module for acoustic feedback. These modules can be rapidly prototyped through 3D printers, laser-cutters, and off-the-shelf components available at a low cost. The Antropo platform can be easily installed on the Franka Emika robot. The added communication channels can be synchronised with the robot's motions to enhance mutual understanding. All hardware CAD design files and software files are released. The platform can be used to study human-like behaviours of cobots and the effects of these behaviours on different aspects of human-robot collaboration. We demonstrate the Antropo platform in an assembly task in which the Franka Emika robot expresses various human-like communicative behaviours via the added communication channels. We also present two industrial applications in which the Antropo platform was customised for the Universal Robots UR16e.

Fast Self-Healing at Room Temperature in Diels-Alder Elastomers.

Despite being primarily categorized as non-autonomous self-healing polymers, we demonstrate the ability of Diels-Alder polymers to heal macroscopic damages at room temperature, resulting in complete restoration of their mechanical properties within a few hours. Moreover, we observe immediate partial recovery, occurring mere minutes after reuniting the fractured surfaces. This fast room-temperature healing is accomplished by employing an off-stoichiometric maleimide-to-furan ratio in the polymer network. Through an extensive investigation of seven Diels-Alder polymers, the influence of crosslink density on self-healing, thermal, and (thermo-)mechanical performance was thoroughly examined. Crosslink density variations were achieved by adjusting the molecular weight of the monomers or utilizing the off-stoichiometric maleimide-to-furan ratio. Quasistatic tensile testing, dynamic mechanical analysis, dynamic rheometry, differential scanning calorimetry, and thermogravimetric analysis were employed to evaluate the individual effects of these parameters on material performance. While lowering the crosslink density in the polymer network via decreasing the off-stoichiometric ratio demonstrated the greatest acceleration of healing, it also led to a slight decrease in (dynamic) mechanical performance. On the other hand, reducing crosslink density using longer monomers resulted in faster healing, albeit to a lesser extent, while maintaining the (dynamic) mechanical performance.

Vitrimeric shape memory polymer-based fingertips for adaptive grasping.

The variability in the shapes and sizes of objects presents a significant challenge for two-finger robotic grippers when it comes to manipulating them. Based on the chemistry of vitrimers (a new class of polymer materials that have dynamic covalent bonds, which allow them to reversibly change their mechanical properties under specific conditions), we present two designs as 3D-printed shape memory polymer-based shape-adaptive fingertips (SMP-SAF). The fingertips have two main properties needed for an effective grasping. First, the ability to adapt their shape to different objects. Second, exhibiting variable rigidity, to lock and retain this new shape without the need for any continuous external triggering system. Our two design strategies are: 1) A curved part, which is suitable for grasping delicate and fragile objects. In this mode and prior to gripping, the SMP-SAFs are straightened by the force of the parallel gripper and are adapted to the object by shape memory activation. 2) A straight part that takes on the form of the objects by contact force with them. This mode is better suited for gripping hard bodies and provides a more straightforward shape programming process. The SMP-SAFs can be programmed by heating them up above glass transition temperature (54°C) via Joule-effect of the integrated electrically conductive wire or by using a heat gun, followed by reshaping by the external forces (without human intervention), and subsequently fixing the new shape upon cooling. As the shape programming process is time-consuming, this technique suits adaptive sorting lines where the variety of objects is not changed from grasp to grasp, but from batch to batch.

Nonlinear Multimaterial Architecture for Greater Soft Material's Toughness and Delaying Damage Propagation.

Designing soft robots that have greater toughness and better resistance to damage propagation while at the same time retaining their properties of compliance is fundamentally important for soft robotics applications. This study's main contribution is proposing a framework for nonlinear multimaterial architectural design of soft structures to increase their toughness and delay damage propagation. What are the limits when combining significantly different materials in one structure that will delay crack propagation while significantly maintaining postdamage toughness? Through this study, we observed that there is a very dynamic interplay when combining significantly different materials in one structure; this interplay could weaken or strengthen the multimaterial structure's toughness. In biological evolutionary terms, the Pangolin, Seashell, and Arapaima have found their answer for deflecting the crack and maintaining strength in their bodies. How does nature put these multimaterial structures together? Our research led us to find that the multimaterial toughness limits depend largely on the following parameters: components' relative morphology, architecture, spatial distribution, surface areas, and Young's Modulus. We found that a linear geometry, when it comes to morphology and/or architecture relative to surface area in multimaterial design, significantly reduces total toughness and fails to delay crack propagation. In contrast, incorporating geometric nonlinearities in both morphology and architecture significantly maintains higher total toughness even after damage, and significantly delays crack propagation. We believe that this study can open the door to further research and ultimately to promising and wide applications in soft robotics.

Assisted damage closure and healing in soft robots by shape memory alloy wires.

Self-healing soft robots show enormous potential to recover functional performance after healing the damages. However, healing in these systems is limited by the recontact of the fracture surfaces. This paper presents for the first time a shape memory alloy (SMA) wire-reinforced soft bending actuator made out of a castor oil-based self-healing polymer, with the incorporated ability to recover from large incisions via shape memory assisted healing. The integrated SMA wires serve three major purposes; (i) Large incisions are closed by contraction of the current-activated SMA wires that are integrated into the chamber. These pull the fracture surfaces into contact, enabling the healing. (ii) The heat generated during the activation of the SMA wires is synergistically exploited for accelerating the healing. (iii) Lastly, during pneumatic actuation, the wires constrain radial expansion and one-side longitudinal extension of the soft chamber, effectuating the desired actuator bending motion. This novel approach of healing is studied via mechanical and ultrasound tests on the specimen level, as well as via bending characterization of the pneumatic robot in multiple damage healing cycles. This technology allows soft robots to become more independent in terms of their self-healing capabilities from human intervention.

Work performance in industry: The impact of mental fatigue and a passive back exoskeleton on work efficiency.

Mental fatigue (MF) is likely to occur in the industrial working population. However, the link between MF and industrial work performance has not been investigated, nor how this interacts with a passive lower back exoskeleton used during industrial work. Therefore, to elucidate its potential effect(s), this study investigated the accuracy of work performance and movement duration through a dual task paradigm and compared results between mentally fatigued volunteers and controls, with and without the exoskeleton. No main effects of MF and the exoskeleton were found. However, when mentally fatigued and wearing the exoskeleton, movement duration significantly increased compared to the baseline condition (ßMF:Exo = 0.17, p = .02, ω2 = .03), suggesting an important interaction between the exoskeleton and one's psychobiological state. Importantly, presented data indicate a negative effect on production efficiency through increased performance time. Further research into the cognitive aspects of industrial work performance and human-exoskeleton interaction is therefore warranted.

Continuous relative phases of walking with an articulated passive ankle-foot prosthesis in individuals with a unilateral transfemoral and transtibial amputation: an explorative case-control study.

BACKGROUND: A mechanical ankle-foot prosthesis (Talaris Demonstrator) was developed to improve prosthetic gait in people with a lower-limb amputation. This study aims to evaluate the Talaris Demonstrator (TD) during level walking by mapping coordination patterns based on the sagittal continuous relative phase (CRP). METHODS: Individuals with a unilateral transtibial amputation, transfemoral amputation and able-bodied individuals completed 6 minutes of treadmill walking in consecutive blocks of 2 minutes at self-selected (SS) speed, 75% SS speed and 125% SS speed. Lower extremity kinematics were captured and hip-knee and knee-ankle CRPs were calculated. Statistical non-parametric mapping was applied and statistical significance was set at 0.05. RESULTS: The hip-knee CRP at 75% SS walking speed with the TD was larger in the amputated limb of participants with a transfemoral amputation compared to able-bodied individuals at the beginning and end of the gait cycle (p = 0.009). In people with a transtibial amputation, the knee-ankle CRP at SS and 125% SS walking speeds with the TD were smaller in the amputated limb at the beginning of the gait cycle compared to able-bodied individuals (p = 0.014 and p = 0.014, respectively). Additionally, no significant differences were found between both prostheses. However, visual interpretation indicates a potential advantage of the TD over the individual's current prosthesis. CONCLUSION: This study provides lower-limb coordination patterns in people with a lower-limb amputation and reveals a possible beneficial effect of the TD over the individuals' current prosthesis. Future research should include a well-sampled investigation of the adaptation process combined with the prolonged effects of the TD.

An Interdisciplinary Tutorial: A Self-Healing Soft Finger with Embedded Sensor.

In the field of soft robotics, knowledge of material science is becoming more and more important. However, many researchers have a background in only one of both domains. To aid the understanding of the other domain, this tutorial describes the complete process from polymer synthesis over fabrication to testing of a soft finger. Enough background is provided during the tutorial such that researchers from both fields can understand and sharpen their knowledge. Self-healing polymers are used in this tutorial, showing that these polymers that were once a specialty, have become accessible for broader use. The use of self-healing polymers allows soft robots to recover from fatal damage, as shown in this tutorial, which increases their lifespan significantly.

Evaluating cognitive and physical work performance: A comparative study of an active and passive industrial back-support exoskeleton.

Occupational back-support exoskeletons, categorized as active or passive, hold promise for mitigating work-related musculoskeletal disorders. However, their impact on combined physical and cognitive aspects of industrial work performance remains inadequately understood, especially regarding potential differences between exoskeleton categories. A randomized, counterbalanced cross-over study was conducted, comparing the active CrayX, passive Paexo Back, and a no exoskeleton condition. A 15-min dual task was used to simulate both cognitive and physical aspects of industrial work performance. Cognitive workload parameters included reaction time, accuracy, and subjective measures. Physical workload included movement duration, segmented in three phases: (1) walking to and grabbing the box, (2) picking up, carrying, and putting down the box, and (3) returning to the starting point. Comfort of both devices was also surveyed. The Paexo significantly increased movement duration in the first segment compared to NoExo (Paexo = 1.55 ± 0.19 s; NoExo = 1.32 ± 0.17 s; p < .01). Moreover, both the Paexo and CrayX increased movement duration for the third segment compared to NoExo (CrayX = 1.70 ± 0.27 s; Paexo = 1.74 ± 0.27 s, NoExo = 1.54 ± 0.23 s; p < .01). No significant impact on cognitive outcomes was observed. Movement Time 2 was not significantly affected by both exoskeletons. Results of the first movement segment suggest the Paexo may hinder trunk bending, favoring the active device for dynamic movements. Both devices may have contributed to a higher workload as the movement duration in the third segment increased compared to NoExo.

Data-driven method for damage localization on soft robotic grippers based on motion dynamics.

Damage detection is one of the critical challenges in operating soft robots in an industrial setting. In repetitive tasks, even a small cut or fatigue can propagate to large damage ceasing the complete operation process. Although research has shown that damage detection can be performed through an embedded sensor network, this approach leads to complicated sensorized systems with additional wiring and equipment, made using complex fabrication processes and often compromising the flexibility of the soft robotic body. Alternatively, in this paper, we proposed a non-invasive approach for damage detection and localization on soft grippers. The essential idea is to track changes in non-linear dynamics of a gripper due to possible damage, where minor changes in material and morphology lead to large differences in the force and torque feedback over time. To test this concept, we developed a classification model based on a bidirectional long short-time memory (biLSTM) network that discovers patterns of dynamics changes in force and torque signals measured at the mounting point. To evaluate this model, we employed a two-fingered Fin Ray gripper and collected data for 43 damage configurations. The experimental results show nearly perfect damage detection accuracy and 97% of its localization. We have also tested the effect of the gripper orientation and the length of time-series data. By shaking the gripper with an optimal roll angle, the localization accuracy can exceed 95% and increase further with additional gripper orientations. The results also show that two periods of the gripper oscillation, i.e., roughly 50 data points, are enough to achieve a reasonable level of damage localization.

Comparison of machine learning and deep learning-based methods for locomotion mode recognition using a single inertial measurement unit.

Locomotion mode recognition provides the prosthesis control with the information on when to switch between different walking modes, whereas the gait phase detection indicates where we are in the gait cycle. But powered prostheses often implement a different control strategy for each locomotion mode to improve the functionality of the prosthesis. Existing studies employed several classical machine learning methods for locomotion mode recognition. However, these methods were less effective for data with complex decision boundaries and resulted in misclassifications of motion recognition. Deep learning-based methods potentially resolve these limitations as it is a special type of machine learning method with more sophistication. Therefore, this study evaluated three deep learning-based models for locomotion mode recognition, namely recurrent neural network (RNN), long short-term memory (LSTM) neural network, and convolutional neural network (CNN), and compared the recognition performance of deep learning models to the machine learning model with random forest classifier (RFC). The models are trained from data of one inertial measurement unit (IMU) placed on the lower shanks of four able-bodied subjects to perform four walking modes, including level ground walking (LW), standing (ST), and stair ascent/stair descent (SA/SD). The results indicated that CNN and LSTM models outperformed other models, and these models were promising for applying locomotion mode recognition in real-time for robotic prostheses.

Humins Blending in Thermoreversible Diels-Alder Networks for Stiffness Tuning and Enhanced Healing Performance for Soft Robotics.

Humins waste valorization is considered to be an essential pathway to improve the economic viability of many biorefinery processes and further promote their circularity by avoiding waste formation. In this research, the incorporation of humins in a Diels-Alder (DA) polymer network based on furan-maleimide thermoreversible crosslinks was studied. A considerable enhancement of the healing efficiency was observed by just healing for 1 h at 60 °C at the expense of a reduction of the material mechanical properties, while the unfilled material showed no healing under the same conditions. Nevertheless, the thermal healing step favored the irreversible humins polycondensation, thus strengthening the material while keeping the enhanced healing performance. Our hypothesis states a synergistic healing mechanism based on humins flowing throughout the damage, followed by thermal humins crosslinking during the healing trigger, together with DA thermoreversible bonds recombination. A multi-material soft robotic gripper was manufactured out of the proposed material, showing not only improved recovery of the functional performance upon healing but also stiffness-tunable features by means of humins thermal crosslinking. For the first time, both damage healing and zone reinforcement for further damage prevention are achieved in a single intrinsic self-healing system.

Transparent Interaction Based Learning for Human-Robot Collaboration.

The number of collaborative robots that perform different tasks in close proximity to humans is increasing. Previous studies showed that enabling non-expert users to program a cobot reduces the cost of robot maintenance and reprogramming. Since this approach is based on an interaction between the cobot and human partners, in this study, we investigate whether making this interaction more transparent can improve the interaction and lead to better performance for non-expert users. To evaluate the proposed methodology, an experiment with 67 participants is conducted. The obtained results show that providing explanation leads to higher performance, in terms of efficiency and efficacy, i.e., the number of times the task is completed without teaching a wrong instruction to the cobot is two times higher when explanations are provided. In addition, providing explanation also increases users' satisfaction and trust in working with the cobot.

Deep learning for biosignal control: insights from basic to real-time methods with recommendations.

Objective.Biosignal control is an interaction modality that allows users to interact with electronic devices by decoding the biological signals emanating from the movements or thoughts of the user. This manner of interaction with devices can enhance the sense of agency for users and enable persons suffering from a paralyzing condition to interact with everyday devices that would otherwise be challenging for them to use. It can also improve control of prosthetic devices and exoskeletons by making the interaction feel more natural and intuitive. However, with the current state of the art, several issues still need to be addressed to reliably decode user intent from biosignals and provide an improved user experience over other interaction modalities. One solution is to leverage advances in deep learning (DL) methods to provide more reliable decoding at the expense of added computational complexity. This scoping review introduces the basic concepts of DL and assists readers in deploying DL methods to a real-time control system that should operate under real-world conditions.Approach.The scope of this review covers any electronic device, but with an emphasis on robotic devices, as this is the most active area of research in biosignal control. We review the literature pertaining to the implementation and evaluation of control systems that incorporate DL to identify the main gaps and issues in the field, and formulate suggestions on how to mitigate them.Main results.The results highlight the main challenges in biosignal control with DL methods. Additionally, we were able to formulate guidelines on the best approach to designing, implementing and evaluating research prototypes that use DL in their biosignal control systems.Significance.This review should assist researchers that are new to the fields of biosignal control and DL in successfully deploying a full biosignal control system. Experts in their respective fields can use this article to identify possible avenues of research that would further advance the development of biosignal control with DL methods.

Processing of Self-Healing Polymers for Soft Robotics.

Soft robots are, due to their softness, inherently safe and adapt well to unstructured environments. However, they are prone to various damage types. Self-healing polymers address this vulnerability. Self-healing soft robots can recover completely from macroscopic damage, extending their lifetime. For developing healable soft robots, various formative and additive manufacturing methods have been exploited to shape self-healing polymers into complex structures. Additionally, several novel manufacturing techniques, noted as (re)assembly binding techniques that are specific to self-healing polymers, have been created. Herein, the wide variety of processing techniques of self-healing polymers for robotics available in the literature is reviewed, and limitations and opportunities discussed thoroughly. Based on defined requirements for soft robots, these techniques are critically compared and validated. A strong focus is drawn to the reversible covalent and (physico)chemical cross-links present in the self-healing polymers that do not only endow healability to the resulting soft robotic components, but are also beneficial in many manufacturing techniques. They solve current obstacles in soft robots, including the formation of robust multi-material parts, recyclability, and stress relaxation. This review bridges two promising research fields, and guides the reader toward selecting a suitable processing method based on a self-healing polymer and the intended soft robotics application.