This paper addresses the research question: “How can a human-robot team achieve co-learning, and interdependence in physically embodied tasks?”
A method has been developed that enables a human-robot team to co-learn the handover of an object from the robot to the human. Five design requirements were composed to address the challenges of human-robot co-learning in physically embodied environments. The method is based on a Q-learning algorithm that was adapted and extended to meet these requirements. An experiment was conducted with six participants. For every human-robot team, each design requirement was qualitatively evaluated. Interdependent co-learning was identified in three of the six teams. The limitation of the design, and how this method can be improved further, was discussed. The method, presented in this paper, demonstrates how human-robot co-learning and interdependence can be enabled in physically embodied tasks.

The increase in online retail demand has stimulated automation in order picking systems, leading to new challenges and opportunities in task assignment and scheduling. In partially automated order picking systems, such challenges and opportunities exist regarding human factors implementation in the job-shop scheduling problem, an optimisation problem essential in operations. Workplace fatigue is a human factor often overlooked in scheduling research and application, despite hurting employees’ well-being and costing U.S. employers up to €127 billion annually. With the opportunities that automation offers, cobotic order picking systems could actively consider human fatigue development, mitigating its negative effects in operation.
This thesis investigates the possibility and potential benefits of fatigue consideration in the job-shop scheduling problem for a partially automated order picking system. We present a new bi-objective mixed integer nonlinear programming problem formulation to represent system constraints and a predictive fatigue model while considering worker fatigue and productivity during schedule optimisation. To put the results of simulated optimisation in perspective, we experimentally validate the fatigue model predictions and fatigue mitigation capabilities of the scheduling approach using heart rate measurements and qualitative fatigue ratings. These experiments occur with employees in a real-life partially automated order picking system.
Our mathematical model can find solutions that the conventional single-objective optimisation approach cannot, allowing fractional energy expenditure distribution improvements more than 4x larger than the decrease in productivity they require in 53% of the considered virtual cases. This is a promising result for fatigue mitigation in operations only by altering operational decision-making. However, the validation experiments show that our predictive fatigue model has an average RMSE of 2.20 kcal/min in estimating energy expenditure rates compared to heart rate measurements while also showing a low correlation. When assessing 10 minute intervals, a time span that fits a scheduling scope, the estimations improve slightly (avg. deviation of -1.85 kcal/min, avg. correlation of 0.17) but still underestimate the measured values. The experiments also show no significant differences in experienced fatigue between existing schedules and those with fatigue mitigation measures applied.
We conclude that the current scheduling formulation is not yet fit for application with a predictive fatigue model. However, real-life operations can benefit from energy expenditure estimation via heart rate measurements and a different approach for implementation is proposed. Research opportunities lie in further fatigue model development and validation, extension to indirect fatigue effects and other human factors, and further development of the mathematical formulation.

Shoulder injuries, prevalent worldwide, often occur from ageing and accidents. In Western countries, these injuries primarily afflict the elderly population, while in rural regions of Bangladesh, Iran, India, and Pakistan, they affect younger individuals who are often the family's primary earners. Due to that, preventing and aiding the recovery of shoulder injuries is crucial. To address this, strain maps with vibrotactile feedback, emerge as a promising solution. However, the feedback system must be affordable, compact, comfortable, user-friendly, easily understood, and portable to suit the local environment. Vibrotactile feedback appears promising but can distract the user from work. Hence, this study seeks to investigate if vibrotactile feedback can be paired with strain maps to guide users in maintaining healthy postures and reducing the risk of shoulder injuries in rural areas, where visual feedback is used as a benchmark. To provide feedback using strain maps, shoulder angles are determined using Python's OpenCV and MediaPipe libraries. PyGame is utilized to display the strain maps, and OpenCV helps delineate boundaries between regions of high and low strain within the shoulder. Visual feedback is integrated into the strain map display, while vibrotactile feedback is delivered through a wearable haptic device. Despite challenges related to axial rotation accuracy and the camera-dependent nature of shoulder angle measurements, user experiments, conducted independently for shoulder elevation and planar elevation, reveal that vibrotactile feedback shows better performance compared to visual feedback. Consequently, this study concludes that vibrotactile feedback has the potential to prevent shoulder injuries with strain maps, but also still needs to improve for future work.

Hand-worn haptic systems must be able to produce high-quality haptics while also being lightweight and energy-efficient. In order to meet market expectations, research is being done into novel drivetrains that may be implemented in hand worn devices. This is being pushed forward by the growing demand in the virtual reality sector, which is placing pressure on the development of new wearable force feedback technology. In order to validate a proof of concept, this thesis proposes a novel force feedback drivetrain with a dual actuator setup-a motor and a brake—that is integrated into a tabletop prototype. Using both a motor and a brake will reduce weight and improve the haptic rendering quality relative to each component when considered separately.

To evaluate the presumptive advantages, a theoretical analysis into the suggested hybrid actuation solution is conducted. The analysis comprises of numerous simulations using a model of the hybrid drivetrain's working principle. Additionally, the analysis is employed to investigate and spot any early-stage defects or undesirable behavior. Consequently, a 3D model was created in order to 3D print a functioning prototype in order to apply the theory in a physical version. The prototype is initially used to validate the simulation model and enforce the findings of the theoretical analysis. Force data is measured when using the prototype and in turn is fed into the simulation model to assess whether the output behaviour is consistent with the prototype. Both the model's competence and the prototype's predictability are assessed using the outcomes.

Finally, an experiment is conducted to both asses the mechanical performance and predictability as well as the participant's perception; 2 virtual environments and 3 actuation modes (motor, brake, and hybrid) were cross-examined and each repeated 6 times, for a total of 36 trials. This experiment ultimately assesses the prototypes validity and determines whether the theoretically assumed benefits are present in the prototype as well. Finally, the participant completes a questionnaire in the form of a 5-point Likert scale to determine if the experiment data complies with the users experience.

The theoretical analysis revealed both reliable behavior and potential problems, such as oscillations in feedback force due to rapid switching between actuator activity. However, the cause of this behaviour was identified and anticipated to be less pronounced during use of the prototype. The prototype was successfully designed and produced, and a comparison to the model revealed consistent findings and, thus, predictable behavior. However, the model occasionally miscalculated the velocity. The experiment also shows that the hybrid drivetrain approach is promising and outperforms the state of the art both in mechanical prowess as well as physical fidelity when providing haptic feedback; the motor fails to render rigid objects and the brake cannot render spring-like objects, while the hybrid approach can render both successfully.

The thesis demonstrates that additional investigation into the suggested drivetrain is supported, despite some obvious limitations. When compared to equally powerful DC motors, the hybrid technique, as employed in the prototype, is capable of portraying the largest variety of haptics, including hard walls, impacts, and stiffness while potentially conserving weight.

The demand for robots in the industry is increasing. Organisations acknowledge the need to innovate in order to keep up with the competition. When robots were first introduced, they took over human tasks if they could execute the task faster than humans. However, this often has a negative effect on the meaningfulness of work. That’s why cobots were introduced, robots that could work together with humans in the same workplace. This provided the industry with many new opportunities. However, as with any new technology, research has to be done to find out how this technology can best be implemented. That’s why FRAIM was founded. They are researching how cobots can be introduced without decreasing the meaningfulness of humans’ work. FRAIM decided to join the Brightsky project, to take their research into practice. However, with the Brightsky project, FRAIM is facing other challenges. Not only does the technical feasibility have to be researched, but also the employees’ needs and other stakeholder values have to be taken into account. A challenge that can be perfectly tackled with a human-centered design approach.

Firstly, through interviewing and stakeholder mapping, the essential stakeholders are identified. These stakeholders have high power or/and high interest in the company and thus contribute to the future of the organisation. Afterwards, the stakeholders are brought together for a co-creation session. In this session, the stakeholders discuss their individual needs and values to identify the contradictions and similarities. Next, a shared vision is created, representing the stakeholders’ view on ‘optimising’ the workflow.
A stakeholder which is often negatively affected by the implementation of robots, is the employee, their needs are often neglected. That’s why this project’s focus is mostly on this stakeholder. During context mapping sessions, the themes of meaningful work are identified. These themes represent the conditions of meaningful work for the employees at KLM Engines.

During the research phase, it became clear that the stakeholders are unaware of the contradictions and similarities of their values. Furthermore, the knowledge and skills needed to understand the possibilities and limitations of implementing robots and their consequences on the stakeholders are lacking.
In the design phase, a co-design is created to tackle this challenge. Co-design is chosen, because it is well-known to be used for bringing together different expertise and for sharing knowledge and skills. Besides, the concept of game design is used. Game design helps to represent a real-life situation in a fun and engaging way and helps the participants to explore different scenarios. Using game design helps to visualise the consequences of robots on the workflow well.

The co-design is tested and evaluated to assess if the co-design meets the design requirements and to discover how the tool can be improved. A final iteration is done to improve the concept.

Lastly, the project is concluded and reflected on. The limitations of the project are discussed and future research and design recommendations are proposed.

Robotic teleoperation has gained popularity in various applications, including the nuclear industry, where the experience and intelligence of a human operator is necessary for making complex decisions that are beyond the autonomy of robots. Human-robot interfaces that help strengthen an operator’s situational awareness without inducing excessive cognitive load are crucial to the success of teleoperation. This paper presents an intuitive visual interface that allows operators to simultaneously control a 6DOF camera platform and a robotic manipulator whilst experiencing the remote environment through a virtual reality head-mounted-display (VR-HMD). The proposed system, Orbital Head-Mounted Display (OHMD), utilizes head rotation tracking to command camera movement in azimuth and elevation directions around a fixation point located at a robot's end-effector. A human factor study was conducted to compare the interface acceptance, perceived workload, and task performance of OHMD with a conventional interface utilizing multiple fixed cameras (Array) and a standard head-mounted display implementation (HMD). Results show that both the OHMD and HMD interfaces significantly improve task performance, reduce perceived workload and increase interface acceptance compared to the Array interface. Whilst orbital head-mounted display excelled in usefulness, the standard HMD interface allowed operators to perform robotic welding tasks significantly faster. Still participants reported they preferred OHMD due to the increased assistance and freedom in viewpoint selection.

To safely deploy legged robots in the real world, it is necessary to provide them with the ability to reliably detect unexpected contacts and accurately estimate the corresponding contact force. Therefore, a collision detection, isolation, and identification pipeline is proposed for a quadrupedal manipulator. An approach based on band-pass filtered forces is presented which accurately detects a collision, and estimates the collision time span. Next, the colliding body link is isolated. Finally, a collision identification method accurately identifies the magnitude and direction of the force. It is robust against model inaccuracies, unmodeled loads and any other potential source of disturbances acting on the robot. The framework is validated using extensive hardware experiments in various scenarios summing up to 416 collisions, including trotting and additional unmodeled load on the robot.

Overtaking on two-lane roads can cause dangerous situations, due to drivers’ errors during the gap acceptance decision. Understanding gap acceptance decisions can help mitigate these situations. Response time (i.e. the time it takes the driver to evaluate the gap and make a decision) has been shown to provide valuable insights into the cognitive processes during gap acceptance decisions in pedestrian crossing and left turn decisions. However, for overtaking, previous studies have not investigated the response time.
I investigated how factors that play a role in gap acceptance decisions during overtaking (namely, size of the gap and the drivers’ velocity) influence the response time. I also studied the change of the drivers' velocity during the decision process. I proposed a novel method to measure the response time in drivers' overtaking decisions, and conducted an experiment to demonstrate the usefulness of the method. 25 participants were presented with multiple overtaking situations in a driving simulator experiment, with varying distance gaps. I analyzed how the probability of gap acceptance varied with the distance gap and participants’ velocity at the start of the overtaking situation using a generalized logistic mixed model. I also analyzed how the response time varied with the distance gap, participants’ velocity and the decision outcome using a linear mixed model. Lastly, I analyzed if the velocity changed between the start and the end of the decision process and whether there was a relation between the decision outcome and the distance gap and the velocity change, using a linear mixed model.
The probability of accepting a gap increased significantly with the distance gap and the velocity of the participant. The response times for rejected gaps were on average 0.7s longer than accepted gaps. The response time increased with the distance gap (42ms per 10m), but decreased with the velocity (-92ms per 1m/s). The velocity changed differently between the decision processes leading to either decisions, with an average difference of 4 m/s.
Using the proposed method, I found that the factors which influence the outcome of the decision, also influence the response time. The dependence of response time on the distance gap and participants’ velocity could be explained by the speed-accuracy tradeoff or the difficulty of the decision. Furthermore, it was shown that the drivers already adapt their velocity during the decision process, instead of after they have made their decision. So, I can conclude that using the proposed method for measuring response time can give insight in the way drivers make gap acceptance decisions during overtaking. My results provide basis for cognitive process models that can help further understand the results and are capable of predicting decision outcomes and response times.

When a trainee is (re)learning a skilled movement, physical guidance from a trainer is crucial. Yet, providing physical cues to guide movements is highly challenging when training is digitally mediated (e.g.remotely). This work demonstrates the utility of pseudo forces generated by a wearable tactile interface for providing non-intrusive movement guidance. First, we developed hardware to generate pseudo forces using asymmetric vibrations, whose frequency and amplitude can be tuned to vary the acceleration of the pseudo forces. Maximum acceleration of 160 m/sec^2 is obtained at a frequency of 40Hz and amplitude of 1. Second, the utility of the generated pseudo forces to provide movement guidance was explored by involving 19 participants in 4 separate experiments: 1) Symmetric and asymmetric vibration comparison, 2) Duration modulation, 3) Amplitude modulation, 4) Frequency Modulation of asymmetric vibration. For every experiment, the elements of movement guidance: direction and joint angular velocity were investigated. Participants perceive directional cues with 96% accuracy (P<0.001), and translate the perceived pseudo forces into directed arm movements, with a uniform joint angular velocity of 14+/-8 degrees/second for the duration of the provided pseudo force. The joint angular velocity of the arm movement changes until 12 degrees/second with frequency. With these findings, we anticipate pseudo forces to be the foundation for remote guidance of human body movements in fields like rehabilitation and sports.

Traffic jams occurring on highways cause increased travel time as well as increased fuel consumption and crashes. Traffic jams without a clear cause, such as an on-ramp or an accident, are called phantom traffic jams and are said to make up 50% of all traffic jams. They are the result of an unstable traffic flow caused by human driving behaviour. Recent studies have shown how automating the longitudinal vehicle motion of only 5% of all cars in the flow can dissipate phantom traffic jams. However, automation introduces new problems, mainly regarding safety when human drivers need to take over the control. This research tries to answer whether phantom traffic jams can be dissolved or prevented using haptic shared control. This means of control keeps the human in the loop and would therefore eliminate the takeover problems while still benefiting from the advantages of automation. 24 participants took part in a driving experiment in a fixed base simulator. In these experiments, we tested haptic shared control against manual control and full automation for longitudinal motion. Results show that traffic jam dissipation performance for haptic shared control lies between manual control and automation. The number of unsafe situations is reduced compared to the automated condition. We conclude that haptic shared control is able to reduce the increased fuel consumption and crashes caused by phantom traffic jams.

Currently, active steering systems are implemented in production vehicles to assist the driver by varying the steering response, where the adaptation of the steering response is either initiated by the vehicle or by the driver. Though studies have shown that these steering systems affect the steering performance positively, the effect on driving behavior and driver acceptance is not taken into account. Moreover, the effect of machine-initiated and driver-initiated steering systems on the driver has yet to be investigated. The aim of this driving simulator study was to examine the effects of machine-initiated and driver-initiated steering systems on driving behavior and driver acceptance. During the experiment, the machine switched between a slow and fast steering response on predetermined locations based on the traffic conditions and road curvature, and the driver could switch the steering response by pressing a mouse-button which was attached to the steering wheel. The expectation was that the preferred steering response would be dependent on the steering task, and that between machine- and driver-initiated steering there would be a trade-off between effort and acceptance. Twenty-four participants drove with a constant velocity on a two-lane road with three sections i.e. overtaking traffic vehicles, driving on a straight road and driving on a curved road. Four conditions were completed in a counter-balanced order i.e. passive slow steering response (PS), passive fast steering response (PF), machine-initiated steering (MI) and driver-initiated steering (DI). A post-experiment questionnaire showed that the participants had a preferred steering response for each of the three sections: between slow and fast for overtaking, slow on the straight road, and fast on the curved road. Furthermore, a lower effort and higher acceptance was achieved with the active steering systems compared to the passive steering systems, where there were no significant differences between MI and DI. For future research, it is recommended to further investigate the use of a range of steering responses and free driving speed.

Conventional steering systems in passenger vehicles have a mechanically fixed steering ratio. The steering sensitivity, defined as the amount of vehicle response to the driver's steering wheel input, remains fixed with changing road environments. Research has shown that driving comfort and safety can be improved when the vehicle's steering sensitivity is adapted to the road curvature profile. Current vehicle models can adapt the vehicle's steering sensitivity based on vehicle's speed and driver's steering wheel angle (i.e variable gear-ratio systems), or on individual selection of driving mode (i.e sport, comfort). It is hypothesised that adaptation of the steering sensitivity based on frequency measures of individual drivers' steering behaviour could improve driving comfort and safety. In a fixed-base driving simulator experiment involving 24 participants, real-time adaptation of steering wheel sensitivity based on individual drivers' steering behaviour was compared to three different fixed steering sensitivity settings on a road with changing road curvature. Here I show that intermittent switching frequency in drivers' steering movements can be used to adapt the vehicle's steering response to a varying road curvature. Significant differences in intermittent switching were found between different road curvature sections and between different steering sensitivity settings. Driver's positional control and comfort ratings did not significantly increase with the steering sensitivity adaptation strategy.

This study proposes an LCA system that provides haptic guidance during lane changes. This system is fully integrated with LKA functionality to provide continuous lateral support during highway driving. Two different system configurations of this LCA are investigated. One is a generalized LCA that provides lane change reference trajectories based on a fixed lane change duration value of 4 seconds. The other is an adaptive LCA that provides personalized lane change reference trajectories through trial-by-trial adaptation to lane change duration of previously driven lane changes. The effects of these systems with respect to mental workload, lateral control performance and user acceptance are investigated. This is observed in an experiment with three different driving sessions for each participant. A manual driving session, a driving session in which the generalized LCA is active and a driving session in which the adaptive LCA is active. The experiments are conducted on a 6 DoF motion-based simulator with 34 participants, driving in a three-lane highway simulation environment with a scripted traffic scenario. To measure mental workload, an auditory cognitive secondary N-back task is introduced. The results show that the introduction of a generalized LCA or adaptive LCA does not have significant influence on mental workload compared to the manual driving session. When the adaptive LCA is introduced, lateral control performance is enhanced compared to the generalized LCA and manual driving. Additionally, user acceptance expressed as subjective usefulness is increased by introducing the adaptive LCA compared to the generalized LCA. Furthermore, inter-driver variability of the lateral control performance during lane changes is reduced by the proposed trial-by-trial adaptive LCA system compared to the generalized LCA system and manual driving.

In conditionally automated driving, drivers should be free to engage in non-driving-related tasks until warned by the system to take over. Over-reliance on automation can lead to dangerous driving behaviour such as engagement in non-related driving tasks during the transition phase and delayed response to take over requests. Take-over time and quality were shown to be influenced by several factors such as involvement in a secondary task, the complexity, and criticality of the driving situation, or the modality and intensity of the take-over request. We found a paucity in research into the effect of post-take-over complexities. Understanding the effect of traffic situation just after transition phase on driver's take-over behavior is crucial for developing highly automated driving systems regarding safety. Therefore, we conducted a driving simulator experiment and tested the influence of post take-over situation complexity on take-over behaviour and driver's trust during automation initiated take-overs. Twenty-four participants drove two conditions, one with high situation complexity and the other one with low situation complexity in a fixed-based driving simulator. Participants were driven with conditionally automated driving and asked to play a game until warned by the automation to take over. The take-over performance was assessed based on safety performance (take over reaction time and no. of collisions) and subjective measures (self-reported perceived trust and technology acceptance). Sub response times in terms of the first gaze and hands on the steering wheel time after the first TOR is prompted, were also analyzed using video annotations. Results showed that situation complexity does not affect take over reaction time, though it significantly affects trust in automation rating. No significant relationship has been found between sub response time and overall take-over reaction time in both situation complexity. Eleven collisions occurred, which indicates that some participants fail to resume control on time. Some participants did not resume control at all and trusted on the automation in both complexities. Higher trust rating in low complexity situation indicates that drivers might develop complacency, resulting in degraded take over ability in unexpected take-overs and delayed response to take-over. This suggests that people develop a wait-and-see attitude and only reclaim control when they feel the need for it. These findings can help design automated technology and policies to ensure the inherent risk in conditionally driving automation can be eliminated.

We are living in an aging society, which is putting an increasingly heavy strain on our healthcare system. As people age many become less mobile, leading to loneliness and a deteriorating health. Subsequently elderly often end up in nursing homes; an experience which is unpleasant as well as expensive. Assisting the elderly with robotic devices can help increase their mobility, therefore reducing healthcare costs and increasing elderly satisfaction. This thesis follows up on the master thesis of R. Berci-Hajnovics, where she sets out to design a novel concept for an assistive, motorized shopping trolley aimed at reducing the physical burden of carrying groceries over uneven terrain. In the current thesis, a control system for the proposed assistive shopping trolley is designed with the goal of attenuating disturbances like road inclination and added mass on the trolley’s dynamic behavior. Predictability of its behavior is considered as well in order to gain the confidence of the elderly user. An analysis of potentially suitable control types for implementation in the assistive shopping trolley has been performed, wherein their characteristics are compared on several different aspects relating to their predictability and disturbance attenuating behavior. Based on this analysis, a controller implementing a so called Disturbance Observer (DOB) is deemed best suited for the job. A DOB uses a dynamic model of the controlled plant to estimate the influence of present disturbances on the plant dynamics. This estimation can consequently be used to produce an opposing force, thereby canceling the effects of the disturbances. Furthermore, a DOB includes a filter which provides additional capabilities, such as robustness to uncertainty within the model and filtering of high-frequency noise in the data to further increase its performance. A DOB is most suitable for the current cause due to its overall adequate performance with respect to the imposed criteria, as well as its relatively high disturbance attenuation accuracy due to its ability to use the plant’s dynamics in making an estimation of the disturbances at play. Following the results of the analysis, this thesis proposes the implementation of a controller including a DOB for application in the assistive shopping trolley. In its design, a model of the trolley dynamics is being used, which provides a basis for the controller’s internal dynamic model. This model is furthermore used as a representation of the trolley plant in a stability analysis of the trolley-controller closed-loop behavior. The model is based on the dynamics of an inverted pendulum on wheels, with an additional constraint on the position of the trolley handle to represent the interaction with the human user. By neglecting minor, nonlinear dynamic effects such as variations in the trolley’s orientation and air drag, the general model is simplified to a linear equation. The controller’s internal model is constructed to represent the trolley’s nominal (desired) behavior, determined to be that of an empty trolley on a flat road, by substituting corresponding nominal values in the parameters of the general model. Subsequently, criteria for robust stability of the trolley-controller’s closed loop behavior are determined using a H_∞-approach; In this approach the system’s behavior is analyzed given the ‘worst-case’ uncertainties. Furthermore, the system’s performance with respect to disturbance estimation and noise filtering is analyzed using its loop transfer function. The insights gained from these analyses are used in the design of the controller and to make recommendations for future work. The obtained controller is implemented in the Robotic Operating System (ROS) framework; an open-source robotics platform which makes use of a decentralized system of processes, simplifying the creation and sharing of complicated software structures across a wide variety of applications. For this purpose, the controller is subdivided into multiple elements, each performing a specific function. These elements, called ‘nodes’, are submitted to a variety of unit tests to verify their proper implementation. The controller’s performance is put to the test in simulation using ROS’s accompanying simulation engine, ‘Gazebo’. The controlled trolley’s acceleration error is compared to that of a regular trolley in several scenario’s, where different types of disturbances are applied. The results show that the controller indeed reduces the effects of disturbances on the trolley’s behavior; the controller is able to recognize the trolley’s changing behavior due to an encountered disturbance and it correctly attenuates its effect by ordering the required compensatory motor torque. More research is required with respect to the effect of uncertainty within the model dynamics to ensure stability and controllability of the trolley system. Moreover, the closed-loop system response should be tested in the presence of various types of noise to determine how well the current results apply when the controller is used in a real-life scenario.

Tele-impedance can increase interaction perfor- mance between a robotic tool and unstructured/unpredictable environment during teleoperation. However, the existing tele- impedance interfaces have several ongoing issues, such as long calibration times and various obstructions for the human oper- ator. In addition, they are all designed to be controlled by the operator’s arms, which can cause difficulties when both arms are used, as in bi-manual teleoperation. To resolve these issues, we designed novel foot-based tele-impedance control method inspired by the human limb stiffness ellipse modulation. The proposed mechanical interface is based around a circular disc and a foot pressure sensor that controls orientation and size/shape of the stiffness ellipsoid, respectively, by the operator’s foot. We evalu- ated the circular disc interface control method in an experimental study with 12 participants, who performed a two-dimensional drilling task in a virtual environment. The results show the ability of the operator in controlling the proposed interface to dynamically adapt to task instruction and environment. In addition, a comparison with low and high constant impedance control demonstrate a superior interaction performance of the proposed interface.

The general approach to generate collision free motion in a constraint environment is to use path planners, which demand a known environment and potentially fail otherwise. Learning from Demonstration (LfD) can be used instead to teach the robot unknown parts of the environment, such as a goal deviation or an unforeseen obstacle. The general approach is to train a model offline and expect it to perform well afterwards. Problems arise however when the model is trained insufficiently or unknown variations have occurred in the environment, which demand for refinement of the model. In online learning the operator is allowed to refine a predicted trajectory during execution time, where the state-of-the-art methods focus on kinaesthetic teaching. The contribution of this research is the development of a teleoperated online learningmmethod, where the operator can make refinements by moving a haptic stylus (Phantom Omni) in the desired direction. By doing this, a force is felt proportional to the magnitude of refinement. After creating such a refined trajectory, it is used to update a condition dependent probabilistic trajectory model. The proof of concept was shown on a 2D example and on a simulated robot that shows that we can adapt an initial model when unknown variations occur and that the method is able to deal with different object positions and initial end effector poses. To show if other people can use the method, a human factors experiment is performed, comparing the developed method against three other methods on how much time it takes to successfully adapt a model (refinement time) and on the perceived workload. Two different parameters are varied, which are the teaching device (stylus or keyboard) and the learning mechanism (online or offline). The expectation was that both online with stylus has the lowest refinement time and workload, but the results show that only online has a significant improvement over offline methods (p = 7.94 × 10^−12 and p = 0.000512 respectively). This is explained by the fact that only small corrections have to be made and in a maximum of three degrees of freedom (DoFs). No significant difference was found between keyboard and stylus (p = 0.755 and p = 0.302 respectively). An explanation for this is that this is task, person and implementation dependent. The recommendations are to evaluate the proof of concept on the real robot and to extend the method with orientation refinement, such that more complex tasks can also be dealtwith. We hypothesize that with these tasks the combination of online with stylus does perform the best.

To make the cooperation within a physical human-robot team as efficient as possible, the team members must be able to co-adapt.
We developed and evaluated a robot that adapts to a human, using an adaptation strategy, in such a way as to guide the co-adaptation to have a positive effect on human task contribution and team performance.
A novel adaptive control algorithm for the robot was designed, estimating and adapting to the human control, using a Nash equilibrium to compute the robot's control inputs. Stability of the controller was theoretically proven, and validated using physical tests.
Two robot adaptation strategies, positive and negative reinforcement, were compared in an experiment in which 18 participants participated. The negative reinforcement adaptation strategy provides assistance on an assist-as-needed basis, whereas the positive reinforcement strategy is designed to intrinsically motivate humans to contribute to the control task.
Results show a significant increase in performance in the negative reinforcement adaptation strategy compared to the positive reinforcement adaptation strategy, whereas both conditions show a significant increase in performance compared to manual control. Results additionally show a significant decrease in both estimated (by the robot) and perceived (by the human) control share in the negative reinforcement adaptation strategy compared to the positive reinforcement adaptation strategy.
In conclusion, to guide the co-adaptation to both increase performance and engage humans to actively contribute to a control task, a robot should be designed to adapt using a positive adaptation strategy.

Programming robots with verbal commands is limited by the capabilities of the utilized natural language parser. A simple natural language parser which can understand only keywords and small phrases may be easy to use, but limited in what it can interpret and convey. Alternatively, a natural language parser which understands more complex commands can be used to convey more nuance, but can be more difficult to use and create. It is unclear when having complex verbal commands available is preferable to having only simple verbal commands available. Here we show that using natural language parsers which understand more complex commands are preferable when teaching a robot, both in terms of user preference and objective metrics such as completion time and accuracy, but only when the task is complex as well. During a preliminary wizard of oz experiment, we observed what types of phrases users use to correct the robot during a pose imitation learning task, in order to create multiple natural language parsers which allowed for different levels of complexity in given verbal feedback. In a follow up experiment, in which 24 users utilized these parsers in a similar task, the users reported finding the more complex ones to be more useful and satisfying to use. Additionally, the more complex parsers also led to a higher objective similarity between the pose that the user wanted to convey and the final attained pose by the robot. However, this last result was only found for poses which required a comparably high effort on part of the user.

This study proposes the creation of a multi-modal feedback system to guide humans towards ergonomic poses. A number of studies have tried to come up with methods where subjects are alerted upon crossing biomechanical or ergonomic thresholds while doing a task but not many have tried to successfully and efficiently guide users to ergonomic positions after having alerted them. Through this study we propose the creation of a multi-modal feedback system comprising of a visual and a speech based audio feedback and hypothesize that the proposed system will lead to a better performance as compared to the other feedback modalities when trying to guide users from one pose to another. During our study we have conducted two sets of experiments to carry out a comparative study between only audio, only visual and the proposed multi-modal feedback system to try and find the modality most effective and successful in guiding humans for pose corrections and a comparative study between two types of speech based audio feedbacks in joint space and end point space to motivate our choice for using the more desired one between the two for our proposed system.
Speech based feedback in joint space came out as the preferred audio feedback due to its ability to allow users to carry out efficient and coordinated inter-joint movements especially in cases of high redundancy whereas the proposed multi-modal feedback system successfully shows its superiority over the other feedback
modalities by showing equivalent results against the benchmark visual feedback when measured objectively and better results when measured subjectively due to its ability to successfully combine the advantages of audio and visual feedback and at the same time, avoid their limitations.