Robotics, Volume 12, Issue 5 (October 2023) – 25 articles

Robotic manipulators can interact with large, heavy objects through whole-arm manipulation. Combined with direct physical interaction between humans and robots, the patient can be anchored in care. However, the complexity of this scenario requires control by a caregiver. We are investigating how such a complex form of manipulation can be controlled by nurses and whether the use of such a system creates physical relief. The use case chosen was washing the back of a patient in the lateral position. The operability of the remote control from the tele-nurse’s point of view, the change in the posture of the nurse on site, the execution times, the evaluation of the cooperation between human and robot, and the evaluation of the system from the nurse’s point of view and from the patient’s point of view were evaluated. The results show that the posture of the worker improved by 11.93% on average, and by a maximum of 26.13%. Ease of use is rated as marginally high. The manipulator is considered helpful. The study shows that remote whole-arm manipulation can anchor bedridden patients in the lateral position and that this system can be operated by nurses and leads to an improvement in working posture. Full article

The ability to predict the maximal performance of an industrial robot executing non-deterministic tasks can improve process productivity through time-based planning and scheduling strategies. These strategies require the configuration and the comparison of a large number of tasks in real time for making a decision; therefore, an efficient task execution time estimation method is required. In this work, we propose the use of neural network models to approximate the task time function of a generic multi-DOF robot; the models are trained using data obtained from sophisticated motion planning algorithms that optimize the shape of the trajectory and the executed motion law, taking into account the kinematic and dynamic model of the robot. For scheduling purposes, we propose to evaluate only the neural network models, thus confining the online use of the motion planning software to the full definition of the actually scheduled task. The proposed neural network model presents a uniform interface and an implementation procedure that is easily adaptable to generic robots and tasks. The paper’s results show that the models are accurate and more efficient than the full planning pipeline, having evaluation times compatible with real-time process optimization. Full article

In this work, we propose two cooperative passivity-based control methods for networks of mechanical systems. By cooperatively synchronizing the end-effector coordinates of the individual agents, we achieve cooperation between systems of different types. The underlying passivity property of our control approaches ensures that cooperation is stable and robust. Neither of the two approaches rely on the modeling information of neighbors, locally, which simplifies the interconnection of applicable systems and makes the approaches modular in their use. Our first approach is a generalized cooperative Interconnection-and-Damping Assignment passivity-based control (IDA-PBC) scheme for networks of fully actuated and underactuated systems. Our approach leverages the definition of end-effector coordinates in existing single-agent IDA-PBC solutions for underactuated systems to satisfy the matching conditions, independently of the cooperative control input. Accordingly, our approach integrates a large set of existing single-agent solutions and facilitates cooperative control between these and fully actuated systems. Our second approach proposes agent outputs composed of their end-effector coordinates and velocities to guarantee cooperative stability for networks of fully actuated systems in the presence of communication delays. We validate both approaches in simulation and experiments. Full article

Reinforcement learning (RL) is explored for motor control of a novel pneumatic-driven soft robot modeled after continuum media with a varying density. This model complies with closed-form Lagrangian dynamics, which fulfills the fundamental structural property of passivity, among others. Then, the question arises of how to synthesize a passivity-based RL model to control the unknown continuum soft robot dynamics to exploit its input–output energy properties advantageously throughout a reward-based neural network controller. Thus, we propose a continuous-time Actor–Critic scheme for tracking tasks of the continuum 3D soft robot subject to Lipschitz disturbances. A reward-based temporal difference leads to learning with a novel discontinuous adaptive mechanism of Critic neural weights. Finally, the reward and integral of the Bellman error approximation reinforce the adaptive mechanism of Actor neural weights. Closed-loop stability is guaranteed in the sense of Lyapunov, which leads to local exponential convergence of tracking errors based on integral sliding modes. Notably, it is assumed that dynamics are unknown, yet the control is continuous and robust. A representative simulation study shows the effectiveness of our proposal for tracking tasks. Full article

Identification is considered a very important procedure, within the control area, to estimate the best-possible approximate model among different designs. Its significance comes from the fact that more than 75% of the cost associated with an advanced control project is aimed at obtaining a precise mathematical modeling. Therefore, in this work, an exhaustive analysis was carried out to determine the appropriate input stimulus for an unknown real system that must be controlled, with the aim of accurately estimating its transfer function (TF) using the empirical identification method (gray-box). The analysis was performed quantitatively by means of three tests: (i) the PID controller step response was evaluated theoretically; (ii) the controller performance was assessed in a Cartesian robot by tracking a trajectory defined through a Gaussian acceleration profile; (iii) the efficiency of the determined input stimulus with the best performance on inferring the TF for the system to be controlled was verified by assessing its operation in a real system, through repeatability tests, utilizing the integral errors. Full article

Artificial perception for robots operating in outdoor natural environments, including forest scenarios, has been the object of a substantial amount of research for decades. Regardless, this has proven to be one of the most difficult research areas in robotics and has yet to be robustly solved. This happens namely due to difficulties in dealing with environmental conditions (trees and relief, weather conditions, dust, smoke, etc.), the visual homogeneity of natural landscapes as opposed to the diversity of natural obstacles to be avoided, and the effect of vibrations or external forces such as wind, among other technical challenges. Consequently, we propose a new survey, describing the current state of the art in artificial perception and sensing for robots in precision forestry. Our goal is to provide a detailed literature review of the past few decades of active research in this field. With this review, we attempted to provide valuable insights into the current scientific outlook and identify necessary advancements in the area. We have found that the introduction of robotics in precision forestry imposes very significant scientific and technological problems in artificial sensing and perception, making this a particularly challenging field with an impact on economics, society, technology, and standards. Based on this analysis, we put forward a roadmap to address the outstanding challenges in its respective scientific and technological landscape, namely the lack of training data for perception models, open software frameworks, robust solutions for multi-robot teams, end-user involvement, use case scenarios, computational resource planning, management solutions to satisfy real-time operation constraints, and systematic field testing. We argue that following this roadmap will allow for robotics in precision forestry to fulfil its considerable potential. Full article

3-XXRRU parallel manipulators (PMs) constitute a family of six-degrees-of-freedom (DOF) PMs with three limbs of type XXRRU, where R and U stand for revolute pair and universal joint, respectively, and XX indicates any actuated two-DOF mechanism that moves the axis of the first R-pair. The members of this family share the fact that they all become particular 3-RRU structures when the actuators are locked. By exploiting this feature, the present paper proposes a general approach, which holds for all the members of this family, to analyze the instantaneous kinematics, workspace, and kinetostatic performances of any 3-XXRRU PM. The results of this study include the identification of singularity conditions without reference to a specific actuation system, the proposal of two specific dimensionless performance indices ranging from 0 to 1, the determination of the optimal actuation system, and the demonstration that 3-XXRRU PMs, when appropriately sized and actuated, possess a broad singularity-free workspace that is also fully isotropic. These findings hold significance in the context of the dimensional synthesis and control of 3-XXRRU PMs. Moreover, when combined with the closed-form solutions for their positional analysis, as demonstrated in a previous publication by the same authors, 3-XXRRU PMs emerge as intriguing alternatives to other six-DOF PMs. The efficacy of the proposed approach is further illustrated through a case study. Full article

Feature extraction plays a crucial role in computer vision and autonomous navigation, offering valuable information for real-time localization and scene understanding. However, although multiple studies investigate keypoint detection and description algorithms in urban and indoor environments, far fewer studies concentrate in unstructured environments. In this study, a multi-task deep learning architecture is developed for keypoint detection and description, focused on poor-featured unstructured and planetary scenes with low or changing illumination. The proposed architecture was trained and evaluated using a training and benchmark dataset with earthy and planetary scenes. Moreover, the trained model was integrated in a visual SLAM (Simultaneous Localization and Maping) system as a feature extraction module, and tested in two feature-poor unstructured areas. Regarding the results, the proposed architecture provides a mAP (mean Average Precision) in a level of 0.95 in terms of keypoint description, outperforming well-known handcrafted algorithms while the proposed SLAM achieved two times lower RMSE error in a poor-featured area with low illumination, compared with ORB-SLAM2. To the best of the authors’ knowledge, this is the first study that investigates the potential of keypoint detection and description through deep learning in unstructured and planetary environments. Full article

Prosthetic hand systems aim at restoring lost functionality in amputees. Manipulation and grasping are the main functions of the human hand, which are provided by skin sensitivity capable of protecting the hand from damage and perceiving the external environment. The present study aims at proposing a novel control strategy which improves the ability of the prosthetic hand to interact with the external environment by fostering the interaction of tactile (forces and slipping) and thermoceptive sensory information and by using them to guarantee grasp stability and improve user safety. The control strategy is based on force control with an internal position loop and slip detection, which is able to manage temperature information thanks to the interaction with objects at different temperatures. This architecture has been tested on a prosthetic hand, i.e., the IH2 Azzurra developed by Prensilia s.r.l, in different temperature and slippage conditions. The prosthetic system successfully performed the grasping tasks by managing the tactile and thermal information simultaneously. In particular, the system is able to guarantee a stable grasp during the execution of the tasks. Additionally, in the presence of an external stimulus (thermal or slippage), the prosthetic hand is able to react and always reacts to the stimulus instantaneously (reaction times ≤ 0.04 s, comparable to the one of the human being), regardless of its nature and in accordance with the control strategy. In this way, the prosthetic device is protected from damaging temperatures, the user is alerted of a dangerous situation and the stability of the grasp is restored in the event of a slip. Full article

The Delta robot is a parallel robot that is over-actuated and has a highly nonlinear dynamic model, which poses a significant challenge to its control design. The inverse kinematics that maps the motor angles to the position of the end effector is highly nonlinear and extremely important for the control design of the Delta robot. It has been experimentally shown that geometry-based inverse kinematics is not accurate enough to capture the dynamics of the Delta robot due to manufacturing component errors, measurement errors, joint flexibility, backlash, friction, etc. To address this issue, we propose a neural network model to approximate the inverse kinematics of the Delta robot with stepper motors. The neural network model is trained with randomly sampled experimental data and implemented on the hardware in an open-loop control for trajectory tracking. Extensive experimental results show that the neural network model achieves excellent performance in terms of the trajectory tracking of the Delta robot under different operation conditions, and outperforms the geometry-based inverse kinematics model. A critical numerical observation indicates that neural networks trained with the specific trajectory data fall short of anticipated performance due to a lack of data. Conversely, neural networks trained on random experimental data capture the rich dynamics of the Delta robot and are quite robust to model uncertainties compared to geometry-based inverse kinematics. Full article

To deploy Unmanned Aerial Vehicles (UAVs) inside heterogeneous GPS-denied confined (potentially unknown) spaces, such as those encountered in mining and Urban Search and Rescue (USAR), requires the enhancement of numerous technologies. Of special interest is for UAVs to identify collision-freeSafe Flight Corridors (SFC${}^{+}$) within highly cluttered convex- and non-convex-shaped environments, which requires UAVs to perform advanced flight maneuvers while exploiting their flying capabilities. Within this paper, a novel auxiliary occupancy checking process that augments traditional 3D flight corridor generation is proposed. The 3D flight corridor is established as a topological structure based on a hand-crafted path either derived from a computer-generated environment or provided by the human operator, which captures humans’ preferences and desired flight intentions for the given space. This corridor is formulated as a series of interconnected overlapping convex polyhedra bounded by the perceived environmental geometries, which facilitates the generation of suitable 3D flight paths/trajectories that avoid local minima within the corridor boundaries. An occupancy check algorithm is employed to reduce the search space needed to identify 3D obstacle-free spaces in which their constructed polyhedron geometries are replaced with alternate convex polyhedra. To assess the feasibility and efficiency of the proposed SFC${}^{+}$ methodology, a comparative study is conducted against the Star-Convex Method (SCM), a prominent algorithm in the field. The results reveal the superiority of the proposed SFC${}^{+}$ methodology in terms of its computational efficiency and reduced search space for UAV maneuvering solutions. Various challenging confined-environment scenarios, each with different obstacle densities (confined scenarios), are utilized to verify the obtained outcomes. Full article

Recent advancements in artificial intelligence have enabled reinforcement learning (RL) agents to exceed human-level performance in various gaming tasks. However, despite the state-of-the-art performance demonstrated by model-free RL algorithms, they suffer from high sample complexity. Hence, it is uncommon to find their applications in robotics, autonomous navigation, and self-driving, as gathering many samples is impractical in real-world hardware systems. Therefore, developing sample-efficient learning algorithms for RL agents is crucial in deploying them in real-world tasks without sacrificing performance. This paper presents an advisor-based learning algorithm, incorporating prior knowledge into the training by modifying the deep deterministic policy gradient algorithm to reduce the sample complexity. Also, we propose an effective method of employing an advisor in data collection to train autonomous navigation agents to maneuver physical platforms, minimizing the risk of collision. We analyze the performance of our methods with the support of simulation and physical experimental setups. Experiments reveal that incorporating an advisor into the training phase significantly reduces the sample complexity without compromising the agent’s performance compared to various benchmark approaches. Also, they show that the advisor’s constant involvement in the data collection process diminishes the agent’s performance, while the limited involvement makes training more effective. Full article

People with disabilities are severely underrepresented in the open labor market. Yet, pursuing a job has a positive impact in many aspects of life. This paper presents a possible approach to improve inclusion by including a robotic manipulator into context-aware Assistive Systems. This expands the assistance possibilities tremendously by adding gesture-based feedback and aid. The system presented is based on the intelligent control system of behavior trees, which—together with a depth camera, specifically designed policies, and a collaborative industrial robotic manipulator—can assist workers with disabilities in the workplace. A developed assistance node generates personalized action sequences. These include different robotic pointing gestures, from simple waving, to precisely indicating the target position of the workpiece during assembly tasks. This paper describes the design challenges and technical implementation of the first Context-Aware Robotic Assistive System. Moreover, an in-field user study in a Sheltered Workshop was performed to verify the concept and developed algorithms. In the assembly task under consideration, almost three times as many parts could be assembled with the developed system than with the baseline condition. In addition, the reactions and statements of the participants showed that the robot was considered and accepted as a tutor. Full article

This review article presents an in-depth examination of research and development in the fields of rehabilitation, assistive technologies, and humanoid robots. It focuses on parallel robots designed for human body joints with three degrees of freedom, specifically the neck, shoulder, wrist, hip, and ankle. A systematic search was conducted across multiple databases, including Scopus, Web of Science, PubMed, IEEE Xplore, ScienceDirect, the Directory of Open Access Journals, and the ASME Journal. This systematic review offers an updated overview of advancements in the field from 2012 to 2023. After applying exclusion criteria, 93 papers were selected for in-depth review. This cohort included 13 articles focusing on the neck joint, 19 on the shoulder joint, 22 on the wrist joint, 9 on the hip joint, and 30 on the ankle joint. The article discusses the timeline and advancements of parallel robots, covering technology readiness levels (TRLs), design, the number of degrees of freedom, kinematics structure, workspace assessment, functional capabilities, performance evaluation methods, and material selection for the development of parallel robotics. It also examines critical technological challenges and future prospects in rehabilitation, assistance, and humanoid robots. Full article

Wire harness manufacturing in the aeronautic sector is highly manual work, with production defined by multiple references and small batches. Although complete automation of the production process is not feasible, a robot-assisted approach could increase the efficiency of the existing production means. This paper presents a novel dual-arm robotic solution for workbench configuration and cable routing during the initial steps of wire harness manufacturing. Based on the CAD information of the wire harness, the proposed framework generates trajectories in real-time to complete the initial manufacturing tasks, dividing automatically the whole job between both robots. The presented approach has been validated in a production environment using different wire harness references, obtaining promising results and metrics. Full article

Geometric errors are the main factors affecting the output accuracy of the parallel spindle head, and it is necessary to perform a sensitivity analysis to extract the critical geometric errors. The traditional sensitivity analysis method analyzes the output position and orientation errors independently, defining multiple sensitivity indices and making it difficult to determine critical geometric errors. In this paper, we propose sensitivity indices that can comprehensively consider position and orientation errors. First, the configuration of the hybrid machine tool is introduced, and the TCP position error model is derived. Then, the tool radius and the effective cutting length are introduced, and the sensitivity indices are defined. After that, the sensitivity analysis of the 3-DOF parallel spindle head is performed using the proposed sensitivity indices, and six critical geometric errors are extracted. The machining accuracy of the parallel spindle head can be greatly improved by improving the critical geometric errors. The proposed sensitivity analysis method can provide important guidance for machine tool accuracy design. Full article

In recent times, the soft robotics field has been attracting significant research focus owing to its high level of manipulation capabilities unlike traditional rigid robots, which gives room for increasing use in other areas. However, compared to traditional rigid gripper robots, being capable of controlling/obtaining overall body stiffness when required is yet to be further explored since soft gripper robots have inherently less-rigid properties. Unlike previous designs with very complex variable-stiffness systems, this paper demonstrates a soft gripper design with minimum system complexity while being capable of varying the stiffness of a continuum soft robotic actuator and proves to have potential applications in gripping objects of various shapes, weights, and sizes. The soft gripper actuator comprises two separate mechanisms: the pneumatic mechanism for bending control and the mechanical structure for stiffness variation by pulling tendons using stepper motors which compresses the actuator, thereby changing the overall stiffness. The pneumatic mechanism was first fabricated and then embedded into another silicon layer during which it was also merged with the mechanical structure for stiffness control. By first pneumatically actuating the actuator which causes bending and then pulling the tendons, we found out that the actuator stiffness value can be increased up to 145% its initial value, and the gripper can grasp and lift a weight of up to 2.075 kg. Full article

Deep reinforcement learning (DRL) has been demonstrated to be effective for several complex decision-making applications, such as autonomous driving and robotics. However, DRL is notoriously limited by its high sample complexity and its lack of stability. Prior knowledge, e.g., as expert demonstrations, is often available but challenging to leverage to mitigate these issues. In this paper, we propose General Reinforced Imitation (GRI), a novel method which combines benefits from exploration and expert data and is straightforward to implement over any off-policy RL algorithm. We make one simplifying hypothesis: expert demonstrations can be seen as perfect data whose underlying policy gets a constant high reward. Based on this assumption, GRI introduces the notion of offline demonstration agent. This agent sends expert data which are processed both concurrently and indistinguishably with the experiences coming from the online RL exploration agent. We show that our approach enables major improvements on camera-based autonomous driving in urban environments. We further validate the GRI method on Mujoco continuous control tasks with different off-policy RL algorithms. Our method ranked first on the CARLA Leaderboard and outperforms World on Rails, the previous state-of-the-art method, by 17%. Full article

Snake-like robots, which have high degrees of freedom and flexibility, can effectively perform an obstacle avoidance motion in a narrow and unstructured space to complete assignments efficiently. However, accurate closed-loop control is difficult to achieve. On the one hand, this is because adding too many sensors to the robot will significantly increase its mass, size, and cost. On the other hand, the more complex structure of the hyper-redundant robot also challenges the more elaborate closed-loop control strategy. For these reasons, a cable-driven snake-like robot, which is compact and low cost, with force transducers and angle sensors, is designed in this article. The simpler and more direct kinematic model is studied, which applies to a widely used kinematics algorithm. Based on the kinematic model, the inverse dynamics are resolved. Finally, this article analyzes the sources of the motion errors and achieves dual-loop control through force-feedback and pose-feedback. The experiment results show that the robot’s structure and dual-loop control strategy function with high accuracy and reliability, meeting the requirements of engineering applications and high-precision control. Full article

In this paper, we propose an in-depth evaluation of the performance of the Archimede rover while traversing rough terrain with loose soil. In order to better analyze this, the reality gap is evaluated when simulating the behavior with an open-source simulator. To this extent, we implement a full model of the rover in the open-source dynamics simulator Gazebo, along with several types of terrains that replicate the experimental conditions. The rover control system is equipped with a kinematics model that allows for driving in different modes. We implement an odometric system aboard the rover, as well as an external optical absolute tracking system as reference. We estimate the drift occurring during driving in different configurations, two types of soil with corresponding wheel geometries. The results show good adherence of the odometry when the rover drives on planar ground; conversely, as expected, a marked influence of slope is seen on wheel drift. The reality gap between simulations and experimental results is kept comparatively small provided that slopes are not present. Full article

In this paper, force sensor signals are classified using a pattern recognition neural network (PRNN). The signals are classified to show if there is a collision or not. In our previous work, the joints positions of a 2-DOF robot were used to estimate the external force sensor signal, which was attached at the robot end-effector, and the external joint torques of this robot based on a multilayer feedforward NN (MLFFNN). In the current work, the estimated force sensor signal and the external joints’ torques from the previous work are used as the inputs to the proposed designed PRNN, and its output is whether a collision is found or not. The designed PRNN is trained using a scaled conjugate gradient backpropagation algorithm and tested and validated using different data from the training one. The results prove that the PRNN is effective in classifying the force signals. Its effectiveness for classifying the collision cases is 92.8%, and for the non-collisions cases is 99.4%. Therefore, the overall efficiency is 99.2%. The same methodology and work are repeated using a PRNN trained using another algorithm, which is the Levenberg–Marquardt (PRNN-LM). The results using this structure prove that the PRNN-LM is also effective in classifying the force signals, and its overall effectiveness is 99.3%, which is slightly higher than the first PRNN. Finally, a comparison of the effectiveness of the proposed PRNN and PRNN-LM with other previous different classifiers is included. This comparison shows the effectiveness of the proposed PRNN and PRNN-LM. Full article

Multirotor Uncrewed Aircraft Systems (UAS), widely known as aerial drones, are increasingly used in various indoor and outdoor applications. For outdoor field deployments, the plethora of UAS rely on Global Navigation Satellite Systems (GNSS) for their localization. However, dense environments and large structures can obscure the signal, resulting in a GNSS-degraded environment. Moreover, outdoor operations depend on weather conditions, and UAS flights are significantly affected by strong winds and possibly stronger wind gusts. This work presents a nonlinear model predictive position controller that uses a disturbance observer to adapt to changing weather conditions and fiducial markers to augment the system’s localization. The developed framework can be easily configured for use in multiple different rigid multirotor platforms. The effectiveness of the proposed system is shown through rigorous experimental work in both the lab and the field. The experimental results demonstrate consistent performance, regardless of the environmental conditions and platform used. Full article

This paper proposes a distance-based formation control strategy with real-time disturbance rejection for omnidirectional mobile robots. The introduced control algorithm is designed such that the leader tracks a desired trajectory while the follower keeps a desired distance and formation angle concerning the leader. In the first step, the evolution of distance and formation angle is obtained from a perturbed second-order dynamic model of the robot, aided by a general proportional integral observer (GPIO), added to estimate unwanted disturbances. Then, the control law is designed for both robots via the active disturbance rejection control (ADRC) methodology, which only depends on the position, distance, and orientation measurements. A numerical simulation compared with a robust controller exhibits the system’s behavior. Furthermore, a set of laboratory experiments is conducted to verify the performance of the proposed control system, where a motion capture system is used as a proof of concept. In this context, this is considered a previous step for further experimentation with onboard sensors. Full article

A resilient robot can recover its original function after partial damage of the system. This paper develops an underactuated resilient robot that utilizes a combination of passive joints, active joints, adjustable links, and passive links. A novel method based on the genetic algorithm was proposed to determine the goal configuration of a partially damaged robot. The novelty of the method lies in the integration of both discrete and continuous variables. This model is illustrated by a 3-DOF robot manipulator in the simulation. Full article

Whilst the use of digital interventions to assist patients with self-management involving embodied conversational agents (ECA) is emerging, the use of such agents to support stroke rehabilitation and recovery is rare. This iTakeCharge project takes inspiration from the evidence-based narrative style self-management intervention for stroke recovery, the ‘Take Charge’ intervention, which has been shown to contribute to significant improvements in disability and quality of life after stroke. We worked with the developers and deliverers of the ‘Take Charge’ intervention tool, clinical stroke researchers and stroke survivors, to adapt the ‘Take Charge’ intervention tool to be delivered by an ECA (i.e., the Taking Charge Intelligent Agent (TaCIA)). TaCIA was co-designed using a three-phased approach: Stage 1: Phase I with the developers and Phase II with people who delivered the original Take Charge intervention to stroke survivors (i.e., facilitators); and Stage 2: Phase III with stroke survivors. This paper reports the results from each of these phases including an evaluation of the resulting ECA. Stage 1: Phase I, where TaCIA V.1 was evaluated by the Take Charge developers, did not build a good working alliance, provide adequate options, or deliver the intended Take Charge outcomes. In particular, the use of answer options and the coaching aspects of TaCIA V.1 were felt to conflict with the intention that Take Charge facilitators would not influence the responses of the patient. In response, in Stage 1: Phase II, TaCIA V.2 incorporated an experiment to determine the value of providing answer options versus free text responses. Take Charge facilitators agreed that allowing an open response concurrently with providing answer options was optimal and determined that working alliance and usability were satisfactory. Finally, in Stage 2: Phase III, TaCIA V.3 was evaluated with eight stroke survivors and was generally well accepted and considered useful. Increased user control, clarification of TaCIA’s role, and other improvements to improve accessibility were suggested. The article concludes with limitations and recommendations for future changes based on stroke survivor feedback. Full article