Quadrupedal robots have great potential for deployment in challenging environments. However, one of the most significant challenges these robots face is maintaining stability on slippery surfaces due to inaccurate friction estimation. This thesis investigates the role of accurate friction estimation in improving locomotion and reducing slip in quadrupedal robots. A Model Predictive Controller (MPC) with a friction-aware constraint update is proposed and evaluated in a simulation environment using a physics-based simulation with Open Dynamics Engine (ODE).

The experimental results demonstrate that incorporating real-time friction measurement and constraint in the MPC framework significantly reduces slip occurrences, decreases energy consumption, and improves overall locomotion stability. Statistical hypothesis tests, including paired t-tests and Bonferroni corrections, confirm the significance of these improvements. The findings suggest that integrating real-time friction estimation into quadrupedal robot controllers can enhance their robustness and reduce the need for explicit slip recovery strategies. Future work should focus on extending this approach to real-world scenarios, incorporating actual friction sensors, and testing in diverse terrains.

The accurate prediction of object softness is crucial in many fields, from agriculture to medical care. Vision-based tactile sensors, which capture high-resolution images of contact interactions, have shown great potential in determining this material property. Many existing approaches, particularly those using end-to-end models, suffer from a 'black-box' problem where it is difficult to understand which features the models use to make their predictions. This lack of transparency makes it challenging to determine and correct errors. To overcome this, this paper shows a data-driven method that can decode information from acquired tactile images to extract pressure distributions and assign softness levels to different objects. A novel approach is explored, integrating a Convolutional Neural Network (CNN) to predict the pressure distribution and a Long Short-Term Memory (LSTM) network to assess material softness. It is demonstrated that the CNN model effectively learns necessary features from the tactile images, enabling precise pressure distribution predictions. Concurrently, the LSTM model analyzes temporal sequences of tactile data, accurately predicting material softness and differentiating ripe from overripe fruits. By utilizing the spatiotemporal pressure distribution, this method improves on existing methods by enabling the efficient use of tactile data and providing additional information that can be used to further enhance the model. This paper can be used as a stepping stone to a more complex system in which robotic control can be implemented based on the sensed material properties, allowing for better control loop mechanisms and expanding the applications of tactile sensing technologies.

Minimally invasive medical procedures often require catheters, endoscopes and other devices to maintain position at a specific site in the body, to cut and remove tissue or for diagnostic purposes. Due to tool force exerted by the surgeon or the natural processes of the body, such as the activity of the heart or the lungs, the inserted device can dislodge or migrate from its intended location. Existing solution focus on stabilizing the tip, following high localized pressure, this can create tissue damage or even perforation. Hence, it can be beneficial to create stabilization over the full length of the catheter. I investigated the use of electroadhesion i.e. using electricity to adhere to the tissue wall. I created a scaled-up, flexible, electroadhesion stabilization proof of concept, consisting of two opposite-charged electrodes in a helical pattern, embedded in silicone rubber. Friction experiments were performed on the catheter proof of concept on two copper half-tube substrates. One of the three flexible scaled-up catheter samples showed an increase in the average dynamic friction coefficient of 10.2% between 0 and 2500 V for the 22 mm diameter substrate. The two other samples showed no or limited increase in friction, attributable to manufacturing differences. Predicting coating thickness is essential, as the coating is the determining factor for the performance of the electroadhesion device. The catheter showed potential, however, improved adhesion performance is required for feasibility on a smaller scale.

Incipient slip detection plays an important role in human and robotic grasping. With the growing use of deep learning in vision-based tactile sensing, the black-box nature of these deep neural networks (DNNs) makes it difficult to analyze, debug, and validate their behavior and learned patterns. To fill this gap, eXplainable AI (XAI) methods have been introduced to shed light into the DNN’s reasoning regarding incipient slip detection. These methods generate saliency maps, highlighting the relevant regions in the input tactile image that resulted in the predicted degree of incipient slip. Temporal difference images have been
used to enhance the visualization of incipient slip and make saliency maps easier for human viewers to understand. Additionally, this research evaluates several XAI methods based on criteria such as high-resolution, smoothness, and faithfulness. The experiment examined 42 samples from the ChromaTouch tactile dataset, focusing on contact interactions with a flat object. The results showed that Poly-CAM satisfies all three criteria by accurately highlighting markers while emphasizing their relative importance in the DNN’s decision-making process. Overall, through visual analysis of saliency maps, our findings confirm that DNNs have successfully learned to localize crucial deformation features for detecting incipient slip.

Creating a lateral force on a finger touching a touch- screen can provide the sensation of a bump in the surface, and generate a pulling sensation, which could enhance the experience of operating a touchscreen. The Ultraloop, a device built to generate such a lateral force on a finger in contact, works by producing flexural traveling waves that push the finger. However, its shape is not simple to manufacture. Moreover, its shape is inconvenient to place in most touch input interfaces, which often appear in flat devices. A flat UltraLoop is proposed providing similar traveling wave force feedback, fitting the form factor of flat human-machine touch interfaces better. Furthermore, the new shape allows for manufacturing from a sheet of aluminum, using a cnc. The flat Ultraloop works by actuating two orthogonal modes at around 38k Hz, with a phase shift of 90 degrees, similarly to the Ultraloop. To validate the performance, force measurements have been conducted showing a maximum lateral force of 0.3N. Also, the influence of the normal force, phase, and finger motion on the lateral force is investigated. Lastly, we present a demo showing application of the flat Ultraloop, simulating a spring and a bump.

Humans perceive a pulling or pushing sensation when subjected to an asymmetric vibration. This so-called pseudo force has great potential to guide human movement. Previous research has exclusively focused on the effect of pseudo forces in open-loop environments, in which the user’s joint angular velocity cannot be corrected. As the latter is essential for providing movement guidance, this paper proposes the first closed-loop system in the field of pseudo forces, using amplitude-modulated pseudo forces as haptic feedback. With this feedback, the user was assisted in moving towards a specific target angular velocity. In a human factors experiment, the amplitude-modulated stimuli were compared to constant-amplitude stimuli. The results showed that amplitude-modulated pseudo forces significantly decreased the error between the user’s and the target angular velocity when continuous movement in the desired direction was achieved. Therefore, the study demonstrated that amplitude-modulated pseudo forces can effectively guide human movement, representing an essential step towards developing a wearable movement guidance device.

In dangerous environments, teleoperation is needed to enable humans to execute tasks remotely. To assist in these tasks, haptic teleoperation systems provide the human operator with the sense of touch of the telerobot. One way to provide this sense of touch is through high-frequency vibration feedback. State-of-the-art solutions generally rely on integrated hardware, which limits their application to specific telerobots and master devices.
The aim of this study is to develop a deployable high-frequency vibration feedback method through an add-on setup. In the presented system, both the vibration recording device and vibration feedback display run on a single microcontroller. Furthermore, all components are small in size and portable by the robotic or human hand.
Spectral analysis of the replicated vibrations shows that the presented system is capable of mimicking textures. To evaluate the effectiveness of the texture imitations, a human factors experiment is conducted. Twenty participants executed a texture identification task for two conditions: a manual condition with direct tactile feedback and a teleoperated condition with tactile feedback displayed by the presented system.
Results show that 75-85% of the textures were correctly identified in the teleoperated condition. These correctness rates are close to the results of the manual condition (96% correct) and outperform the chances of random guessing by a factor three. In the teleoperated condition, participants took on average 67% longer than in the manual feedback condition.
Based on these results, it is concluded that the presented add-on system enables humans to accurately feel high-frequency vibrations in teleoperation.

While defusing a bomb or performing a rescue mission with a teleoperated robot, grasping various objects is crucial. Despite being a routine activity, remote grasping is still challenging. It is difficult to apply an adequate grip force to avoid slippage and damage to an object. An additional challenge is controlling both motion and force at the same time during remote robot control (teleoperation). Therefore, this research presents a teleoperated semi-autonomous controller which assists the user with remote grasping by relieving the user from controlling the grip force. Our design enables the user (1) to control the position of the remote gripper while (2) the system controls the grip force autonomously. When the user grasps an object, the semi-autonomous controller maintains the grip force based on tactile feedback to prevent object slippage. For tactile feedback, our system uses a tactile sensor that can detect incipient slippage from deformations at the location of the contact. With two experiments, we show that the system can maintain an adequate grip force while being robust to external perturbations and input changes. Since this controller stably grasps objects while the user maintains control over the position of the remote robot, our method relieves the user and prevents object slippage

In the field of soft robotics, rigid joints and links are replaced by soft, deformable elements, This causes soft continuum robot arms to excel in unpredictable environments, but to face challenges during control and shape reconstruction. The sensing ability present in octopus suckers provides inspiration for solutions. Octopuses employ their suckers not only to strengthen their grasp but also as tactile sensors to control the shape and position of their soft arms. This has motivated researchers to integrate artificial sensorized suckers in soft continuum robot arms Although various sensorized suckers have already been developed, their employed sensing methods tend to be low in resolution and are often poorly embedded into the overall sucker architecture. In this work, these limits are overcome by presenting an octopus-inspired suction cup with integrated high-resolution tactile sensing abilities. This is achieved by utilizing the Chromatouch Principle, which relies on embedding colored markers in the suction cup membrane. Tracking these markers with a camera produced tactile images containing useful information about forces, deformations and interactions with objects. Fabrication with multi-material additive manufacturing enabled direct integration of these markers into the suction cup membranes. We demonstrated the design’s basic functionality by conducting pull-off and pickup tests. The design exhibited a normal pull-off force of 9.53 N and a shear pull-off force of 5.28 N. It was also able to successfully pick up both flat and curved objects. The sensing ability was showcased by training a Convolutional Neural Network to learn the relationship between the camera images and the orientation of the suction cup with respect to a touching substrate. Using a spherical coordinate system, the orientation could be predicted with an error of less than 2 degrees for latitude and less than 9 degrees for longitude. This performance was validated by using the trained network to successfully correct the orientation when picking up objects under an angle. For a single suction cup, this ability can be utilized to correct the orientation and achieve perpendicular contact with an object, crucial for achieving a seal. On a larger scale, the integration of multiple suction cups in soft continuum robot arms has the potential to form a representation of the arm shape as a whole. It can thereby contribute to overcoming the control challenges faced in the field of soft robotics.

When we manipulate objects in our day-to-day life, we perceive information on the object via force feedback that we sense with our sensorimotor system. However, in virtual reality, we lack these forces, which makes it more challenging to interact with the digital world. Wearables, such as hand exoskeletons, can provide force feedback in VR. Nonetheless, as the devices’ actuation or brake system is often bulky or heavy, users typically do not enjoy wearing them. In this study, we investigate the potential of a new friction-based mechanism that can address the existing issues. Our design adopts ultrasonic vibrations to generate a squeeze film, which is a well-studied phenomenon to decrease friction significantly. In literature, the phenomenon has only been investigated with vibrations from one side, with the goal of friction reduction. However, we show that by adding a vibrating surface opposite to the original one, we can extend the possibilities of squeeze films, enabling us to both decrease and increase friction. Since ultrasonic transducers can be miniaturized, our mechanism brings us closer to solving the size and weight issues of existing devices.

It is impossible to imagine modern day interaction with technology without the use of touchscreens. It is a go-to interface to use for many applications, because of the high stimuli-response compatibility and adaptability of the graphical user interface. But the haptic feedback one would have with physical buttons and dials, is lost with the use of touchscreens. High potential to improve the interaction with high resolution haptic feedback is often ignored. In this paper, the use of a haptic pseudo-potential field rendering method on a friction modulated touchscreen is proposed. With this method, the user is assisted in moving towards a target by lowering the friction and impeded in moving away by increasing the friction coefficient. In a human factors experiment, this rendering method is compared to a position-based friction modulation method. Subjects are instructed to find a target path, based on the haptic feedback. The results show that the position-based rendering method has a higher hit-rate and lower movement times. This demonstrates that the pseudo-potential field method is difficult to perceive, however it is expected that advancements in rendering larger friction coefficient ranges or even active lateral force feedback will improve this rendering method.

Our remarkable sense of touch provides us the feedback that is crucial for successfully manipulating a wide range of objects.
The unconscious synergy between touch and the precision grip is particularly astonishing.
During precision manipulation, humans constantly control their grip force to maintain a safety margin of approximately 25 percent above the minimum force required to prevent held objects from slipping.
The ability to accurately control this safety margin heavily relies on tactile feedback founded on sensed deformations of our fingertips.
Previous studies have demonstrated that, by using this feedback, humans even manage to maintain this safety margin independently of the weight or friction of a lifted object, and when the weight of a held object is perturbed.
However, it is still unknown whether the sense of touch can help us to maintain this safety margin when the friction of a statically held object is perturbed.
As previous methods could not deliver these friction perturbations, we demonstrated the viability of a new friction perturbation method that we employed to fill this knowledge gap.
Here we show that humans in fact do not adapt their grip force in response to an abrupt increase of friction, but do increase their grip force in response to an abrupt decrease of friction.
The asymmetry of these grip adaptations is consistent with current hypotheses on the limitations of our sense of friction.
Our results support the existence of the hypothesized inability of our sense of touch to directly sense an increase of friction.
These findings can help to enhance the haptic interaction between humans and machines, and may inspire the design of an artificial sense of touch that can greatly improve the manipulation dexterity of robotic grippers.

Vibrotactile wearable devices are a non-intrusive and inexpensive means to provide haptic feedback directly on the user’s skin. These devices utilize one or multiple vibrotactile actuators to generate vibrations across the skin and into the tissue. Combining these vibrations in amplitude can create the illusion of a funneled sensation on the skin at another location than at the actual sites of stimulation. This allows for the placement of virtual actuators on the skin, such that fewer actuators need to be deployed. However, the illusion does not take into account that the waves originating from the actuator attenuate and disperse due to the viscoelastic properties of the skin. We hypothesize that this diffusion of the elastic energy in the skin is affecting the perception of this illusion. Therefore, if we correct for the wave propagation speed, and temporally focus the stimulation, we hypothesized that the specificity of the stimulation on the skin could be drastically improved. In this paper, a novel technique, which is named the inverse filter technique, was introduced that enables to focus the amplitude, frequency and phase of vibrations to one location while cancelling them at the remaining nearby positions. We developed a wearable device for the volar surface of the forearm on which we could independently control arbitrary waveforms at any position between a set of four physical actuators. A human-subject study found that the performance in terms of localization confidence was improved significantly, whereas the precision and accuracy of the task did not improve compared to when we did not correct for the wave attenuation and dispersion. These results show that focusing waves towards a target location has a direct influence on our confidence of localizing vibrotactile stimuli on the arm. Therefore, we anticipate that our findings can benefit industries interested in including localized vibrotactile feedback on the human body surface.

Tactile sensing provides crucial information about the stability of a grasped object by a robotic gripper. Tactile feedback can be used to predict slip, allowing for timely response to perturbations and to avoid dropping objects. Tactile sensors, included in robotic grippers, measure vibrations, strain or shearing forces which are produced by the movement of the grasped object. With sufficient spatial resolution, tactile sensors can even classify slip or estimate the 3d force displacement field. However, current tactile sensors fail to preemptively detect slippage, requiring fast reaction times during applications in real-time control. Here we show a perception framework that can predict slippage before it occurs by estimating the frictional safety margin. The safety margin indicates the margin to the frictional strength of a grasp, which decreases for reduced friction or increased load force. An accurate safety margin estimate allows for more efficient robot grip force control while providing robustness against object uncertainty and frictional conditions. We developed a high resolution tactile sensor, on which we trained a convolutional neural network to learn the relationship between tactile images and the safety margin. The network’s performance is evaluated on unseen test data, showing robustness to variations in environmental conditions. The results demonstrate that the tactile images contain the information needed to produce accurate safety margin estimates. These estimates can be used for control up to 20% of the minimum required grip force, mimicking human grasping behavior. This approach can drive new grasp control methods and enable robotic grasping of fragile objects in highly dynamic environments. Applications can be found in harvesting, parcel sorting, or improving human-robot interaction.

Conventional robots adopt wheels or robotic limbs for locomotion. Wheels are simple to control but not suitable for irregular terrains. On the other hand, robotic feet can overcome a wider variety of surfaces but are less desirable due to their complex design and control system. In nature, we find that invertebrate animals, like snails, can go anywhere without needing the complex control system a rigid robot needs. Instead, they move forward by sending travelling deformations along their soft bodies. Inspired by these animals, we present a soft robot that uses a sequence of deformations for locomotion. This sequence of deformations is driven by a row of vibrating actuators that vibrate with the same frequency but a phase shift between each consecutive actuator. The metachronal wave arising from this vibration pattern offers efficient locomotion due to the continuous movement of the robot. The top speed achieved in this research was 5 mm/s for both forward and backward locomotion. Our study shows how the robot’s locomotion capabilities are affected by its material and how the robot's velocity depends on the mechanical design and the properties of the metachronal wave and provides a next step into soft robots that can efficiently move almost anywhere.

Manipulating soft and fragile objects is a challenging task in robotic grasping. The key challenge for robotic grasping is to exert enough grip force to prevent slipping while being gentle enough to prevent damage to an object. Existing grippers used for processes like automatic harvesting of fruits, either apply excessive grip force leading to object damage or react to slip resulting in object release from the gripper. The aim of this study is to develop a grip force controller that uses tactile feedback to maintain a constant frictional safety margin over the minimum required grip force, called Safety Margin Control. Tactile sensors can provide information on friction, which is used to predict slip. An optical tactile sensor is modeled and used in simulations where Safety Margin Control regulates the grip force during interaction with various virtual objects. The deformation of the sensor’s soft viscoelastic membrane is described by local frictional behavior and used to estimate the safety margin. The desired safety margin is set to 30%, based on comparison to the way humans control grip force in their fingertips. The desired value can be tuned to favor release over damage and vice versa. Safety Margin Control is compared to two baseline controllers: React To Slip and Conservative Control. The performance is evaluated based on maximum pressure and total lateral displacement of the object relative to the sensor. Safety Margin Control results in a pressure decrease of 44% on average compared to Conservative Control, and no significant pressure change was observed compared to React To Slip. The total lateral displacement for Safety Margin Control is 0 mm, as opposed to 1.3 mm for React To Slip. Safety Margin Control provides a way forward for automated harvesting as the pressure exerted on an object can be reduced while no slip occurs.

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.

Tendon-driven soft exosuits have gained much attention in recent years due to their ability to support people with mobility problems during daily life activities. A task that is particularly assisted by such robotic systems, due to its challenging nature, is stair climbing. However, assisting foot clearance during the swing phase of stair climbing with a tendon-driven system that actively supports hip and knee flexion has not been investigated.
In the current study, we built a tendon-driven system to experimentally identify the foot clearance, the joint moments, the tendon forces, and the tendon velocities, that the system should provide to lift the foot above a stair-step. We experimented with able-bodied subjects that, while standing still, were instructed to keep their right leg relaxed to mimic muscle weakness. With this experimental approach, we showed how to systematically identify the basic requirements for developing an assistive device.
The system provided up to 40.1 Nm at the hip and up to 23.2 Nm at the knee when simultaneously flexing both joints, while not being constrained by any space limitations. When modified to mimic a wearable, portable exosuit, that does not protrude more than 10 cm from the body, the setup provided an average foot clearance of 17 cm with thigh tendon forces up to 455 N and shank tendon forces up to 50 N. Consequently, exosuits that support hip and knee flexion should exploit the low required shank tendon forces and the possibility of simultaneously flexing the hip and knee joints to more efficiently assist foot clearance during stair climbing.

Vibration energy harvesters have been proposed as a solution to increase the lifetime of wireless and portable medical devices. One example of implantable medical devices for which energy harvesters could be interesting, are pacemakers. With a lifespan of about 6 to 12 years, the battery must be replaced after this period of time. Using an energy harvester instead of a battery is therefore seen as an interesting alternative. However, the human heart rate is usually between 0.6-2Hz and consists of low acceleration peaks (<1g). The use of resonance at low frequency is extremely difficult, especially when the motion amplitude is larger than the device itself. A solution is sought in the non-resonant bistable energy harvesters. When enough force is applied to overcome the potential energy barrier, snap through motion is induced, resulting in a significant increase in power output. However, large threshold accelerations are limiting the usability of these systems. Therefore, stiffness compensation is required. A prototype was fabricated in which buckled flexures were used to add negative stiffness to a piezoelectric cantilever, resulting in a stiffness compensated bistable energy harvester suitable for energy harvesting from low frequency and low force excitations. The dynamical behaviour and practical performance of the prototype was studied in relation to a heartbeat, sawtooth wave and sine waves. The output power of the non-resonant prototype was compared to a resonant device, which in all cases showed that the non-resonant prototype outperformed the resonant device. This shows that stiffness compensated bistable energy harvesters can be used in order to make energy harvesting for low force and low frequency excitations, such as a heartbeat, possible.

An integral part of creating compelling and useful immersive virtual environments is the presence of haptic feedback, where both kinesthetic and cutaneous feedback convey unique information critical to object handling and manipulation. State-of-the-art haptic gloves are broadening the workspace and degrees of freedom for hand motion in VR, adding force and vibrotactile feedback to the user's hand and fingers. The effect of each feedback modality to task performance remains unclear. In the current proof-of-concept study we added high-frequency vibrotactile actuators onto a resistive force-feedback glove, to illustrate how our proposed analysis method might contribute to the field of human factors studies. We aim to identify improvements to a haptic glove by isolating the contribution of both haptic feedback modalities, that is the force and vibrotactile feedback. Haptic gloves are used to train the fine motor control skills for assembly operations, allowing the users to grasp objects in an natural manner. By tracking the Finger Indentation -- the distance that a fingertip travels into the virtual surface of the grasped object -- we can capture the grasping realism achieved by each haptic condition. This approach allows us to explore the differences between expected physical world behavior and corresponding behavior in a virtual environment. To determine if our proposed metric provides insights into human behavior, we perform a within-subject comparison of a simplified assembly task in virtual reality with 3 participants. Participants are asked to perform a pick-and-place task for 4 conditions: no feedback, vibrotactile feedback, force feedback, and combined feedback. Performance is judged by analysing task completion time, finger indentation, and subjective measurements. The results of this proof-of-concept study suggest that the combined feedback condition performs better in terms of lower thumb indentations and higher usefulness and satisfactory scores than the no feedback and vibrotactile feedback conditions. It is notable that the vibrotactile feedback condition did not seem to out-perform the no-feedback condition. Regardless, the finger indentation metric seems to be a valid method to compare feedback modalities for grasping performance in a full-sized human factors study.