Search

Combining functional soft materials with electrical impedance tomography is a promising method for developing continuum sensorized soft robotic skins with high resolutions. However, reconstructing the tactile stimuli from surface electrode measurements is a challenging ill-posed modelling problem, with FEM and analytic models facing a reality gap. To counter this, we propose and demonstrate a model-free superposition method which uses small amounts of real-world data to develop deformation maps of a soft robotic skin made from a self-healing ionically conductive hydrogel, the properties of which are affected by temperature, humidity, and damage. We demonstrate how this method outperforms a traditional neural network for small datasets, obtaining an average resolution of 12.1 mm over a 170 mm circular skin. Additionally, we explore how this resolution varies over a series of 15,000 consecutive presses, during which damages are continuously propagated. Finally, we demonstrate applications for functional robotic skins: damage detection/localization, environmental monitoring, and multi-touch recognition - all using the same sensing material.

International audience; The intrinsic compliance of soft robots provides safety, a natural adaptation to its environment, allows to absorb shocks, and protects them against mechanical impacts. However, a literature study shows that the soft polymers used for their construction are susceptible to various types of damage, including fatigue, overloads, interfacial debonding and cuts, tears and perforations by sharp objects. An economic and ecological solution is to construct future soft robotic systems out of self-healing polymers, incorporating the ability to heal damage. This review paper proposes criteria to evaluate the potential of a self-healing polymer to be used in soft robotic applications. Based on these soft robotics requirements and on defined performance parameters of the materials, linked to the mechanical and healing properties, the different types of selfhealing polymers already available in literature are critically assessed and compared. In addition to a description of the state of the art on self-healing soft robotics, the paper discusses the driving forces and limitations to spur the interdisciplinary combination between self-healing polymer science and soft robotics.

Soft strain sensors have been widely used for the development of electronic skins both for robotic and wearable applications. To sense contact location on a wide surface, the standard methodology consists of square grids of strain fibers that are able to detect single contact points but fail to detect multiple ones simultaneously. To avoid such a problem, state-of-the-art technologies implement sequential sampling that isolates each sensing node, but at the cost of a lower sampling rate. This theoretical study proposes a design methodology for multi-touch detection for parallel processed grid-based strain sensors. The fundamental idea is to add diagonal grids of varying orientations on top of the standard architecture to achieve multi-touch detection. The maximum number of detectable points and the number of required strain fibers and the overall geometry of the sensor are studied along with the error introduced when trying to sense more contact points than designed for. Overall, compared with state-of-the-art design methodologies, our work provides a guideline for more efficient grid-based architectures that are able to simultaneously detect up to a fixed finite number of contact points

The ability to 3D print soft materials with integrated strain sensors enables significant flexibility for the design and fabrication of soft robots. Hydrogels provide an interesting alternative to traditional soft robot materials, allowing for more varied fabrication techniques. In this work, we investigate the 3D printing of a gelatin-glycerol hydrogel, where transglutaminase is used to catalyse the crosslinking of the hydrogel such that its material properties can be controlled for 3D printing. By including electron-conductive elements (aqueous carbon black) in the hydrogel we can create highly flexible and linear soft strain sensors. We present a first investigation into adapting a desktop 3D printer and optimizing its control parameters to fabricate sensorized 2D and 3D structures which can undergo >300% strain and show a response to strain which is highly linear and synchronous. To demonstrate the capabilities of this material and fabrication approach, we produce some example 2D and 3D structures and show their sensing capabilities. © 2016 IEEE.

Autonomous dishwasher loading is a benchmark problem in robotics that highlights the challenges of robotic perception, planning, and manipulation in an unstructured environment. Current approaches resort to a specialized solution, however, these technologies are not viable in a domestic setting. Learning‐based solutions seem promising for a general purpose solutions; however, they require large amounts of catered data to be applied in real‐world scenarios. This article presents a novel learning‐based solution without a training phase using pre‐trained object detection networks. By developing a perception, planning, and manipulation framework around an off‐the‐shelf object detection network, we are able to develop robust pick‐and‐place solutions that are easy to develop and general purpose requiring only a RGB feedback and a pinch gripper. Analysis of a real‐world canteen tray data is first performed and used for developing our in‐lab experimental setup. Our results obtained from real‐world scenarios indicate that such approaches are highly desirable for plug‐and‐play domestic applications with limited calibration. All the associated data and code of this work are shared in a public repository.

Modeling of soft robots is typically performed at the static level or at a second-order fully dynamic level. Controllers developed upon these models have several advantages and disadvantages. Static controllers, based on the kinematic relations tend to be the easiest to develop, but by sacrificing accuracy, efficiency and the natural dynamics. Controllers developed using second-order dynamic models tend to be computationally expensive, but allow optimal control. Here we propose that the dynamic model of a soft robot can be reduced to first-order dynamical equation owing to their high damping and low inertial properties, as typically observed in nature, with minimal loss in accuracy. This paper investigates the validity of this assumption and the advantages it provides to the modeling and control of soft robots. Our results demonstrate that this model approximation is a powerful tool for developing closed-loop task-space dynamic controllers for soft robots by simplifying the planning and sensory feedback process with minimal effects on the controller accuracy.

The ability to sense compliant interactions through internal state measurements improves adaptability and robustness of robotic grasping and manipulation, without interfering with any carefully designed passive dynamics. This work presents a collocated sensing and actuation system using a compliant tendon driven hand and its accompanying model. Using a simple vision-based measurement device the approach is able to estimate its internal state and applied external forces. The camera-based sensor uses off-the-shelf components and has the ability to measure many more than 15 tendons simultaneously, giving a low-cost, scale-able, high-bandwidth sensing solution. The collocated configuration provides a framework for developing stable feedback controllers. Our results show that the system can estimate change in posture with $\lt 10$% error with the potential to estimate contact forces using the same scheme.

With advances in the field of robotic manipulation, sensing and machine learning, robotic chefs are expected to become prevalent in our kitchens and restaurants. Robotic chefs are envisioned to replicate human skills in order to reduce the burden of the cooking process. However, the potential of robots as a means to enhance the dining experience is unrecognised. This article introduces the concept of food quality optimization and its challenges with an automated omelette cooking robotic system. The design and control of the robotic system that uses general kitchen tools is presented first. Next, we investigate new optimization strategies for improving subjective food quality rating, a problem challenging because of the qualitative nature of the objective and strongly constrained number of function evaluations possible. Our results show that through appropriate design of the optimization routine using Batch Bayesian Optimization, improvements in the subjective evaluation of food quality can be achieved reliably, with very few trials and with the ability for bulk optimization. This study paves the way towards a broader vision of personalized food for taste-and-nutrition and transferable recipes. © 2016 IEEE.

Soft robotic grippers are shown to be high effective for grasping unstructured objects with simple sensing and control strategies. However, they are still limited by their speed, sensing capabilities and actuation mechanism. Hence, their usage have been restricted in highly dynamic grasping tasks. This paper presents a soft robotic gripper with tunable bistable properties for sensor-less dynamic grasping. The bistable mechanism allows us to store arbitrarily large strain energy in the soft system which is then released upon contact. The mechanism also provides flexibility on the type of actuation mechanism as the grasping and sensing phase is completely passive. Theoretical background behind the mechanism is presented with finite element analysis to provide insights into design parameters. Finally, we experimentally demonstrate sensor-less dynamic grasping of an unknown object within 0.02 seconds, including the time to sense and actuate.

Control of robots has largely been based on the assumption of a fixed morphology. Accordingly, robot designs have been stationary in time, except for the case of modular robots. Any drastic change in morphology, hence, requires a remodelling of the controller. This work takes inspiration from developmental robotics to present a piecewise morphology-controller growth/adaptation strategy that facilitates fast and reliable control adaptation to growing robots. We demonstrate our methodology on a simple 3 degree of freedom walking robot with adjustable foot lengths and with varying inertial conditions. Our results show not only the effectiveness and reliability of the piecewise morphology controller co-adaptation (PMCCA) strategy, but also highlight the need for morphological adaptation as a robot design strategy.

Soft robots have garnered interest for real-world applications because of their intrinsic safety embedded at the material level. These robots use deformable materials capable of shape and behavioral changes and allow conformable physical contact for manipulation. Yet, with the introduction of soft and stretchable materials to robotic systems comes a myriad of challenges for sensor integration, including multimodal sensing capable of stretching, embedment of high-resolution but large-area sensor arrays, and sensor fusion with an increasing volume of data. This Review explores the emerging confluence of e-skins and machine learning, with a focus on how roboticists can combine recent developments from the two fields to build autonomous, deployable soft robots, integrated with capabilities for informative touch and proprioception to stand up to the challenges of real-world environments.

Recent electrophysiological observations related to saccadic eye movements in rhesus monkeys, suggest a prediction of the sensory consequences of movement in the Purkinje cell layer of the cerebellar oculomotor vermis (OMV). A definite encoding of real-time motion of the eye has been observed in simple-spike responses of the combined burst-pause Purkinje cell populations, organized based upon their complex-spike directional tuning. However, the underlying control mechanisms that could lead to such action encoding are still unclear. We propose a saccade control model, with emphasis on the structure of the OMV and its interaction with the extra-cerebellar components. In the simulated bilateral organization of the OMV, each caudal fastigial nucleus is arranged to receive incoming projections from combined burst-pause Purkinje cell populations. The OMV, through the caudal fastigial nuclei, interacts with the brainstem to provide adaptive saccade gain corrections that minimize the visual error in reaching a given target location. The simulation results corroborate the experimental Purkinje cell population activity patterns and their relation with saccade kinematic metrics. The Purkinje layer activity that emerges from the proposed organization, precisely predicted the speed of the eye at different target eccentricities. Simulated granular layer activity suggests no separate dynamics with respect to shaping the bilateral Purkine layer activity. We further examine the validity of the simulated OMV in maintaining the accuracy of saccadic eye movements in the presence of signal dependent variabilities, that can occur in extra-cerebellar pathways.

The complex motion abilities of the Octopus vulgaris have been an intriguing research topic for biologists and roboticists alike. Various studies have been conducted on the underlying control architectures employed by these high dimensional biological organisms. Researchers have attempted to replicate these architectures on robotic systems. Contrary to previous approaches, this study focuses on a robotic system, which is only morphologically similar to the Octopus vulgaris, and how it would behave under different control policies. This sheds light on the underlying optimality principles that these biological systems employ. Open loop control policies are obtained through a trajectory optimization method on a learned forward dynamic model. The motion patterns emerging from variations in morphology and environment were then derived to study the role of the body and environment. Results show that for the specific case of dynamic reaching with a soft appendage, the invariance in motion profile is a fundamental constraint imposed by the morphology and environment, independent from the controller. This suggests how morphological design can simplify stable control even for highly dimensional nonlinear dynamical systems and can provide insights into design of new soft robotic mechanisms.

Soft robots are diverse time varying dynamical systems. Hence models describing their behavior have to constantly adapt to accommodate these changes. This is a challenging problem due to the complexities involved with modelling a soft robot and the diverse array of available modelling approaches. Adaptive controllers that can accommodate for model uncertainties are a viable option in this case. This paper presents a cerebellum-inspired adaptive kinematic controller. The adaptive controller is built on top of an approximate inverse kinematic model. The proposed architecture guarantees error convergence without a priori knowledge about the approximate inverse kinematic controller and learns using only task space error information. Simulation results show how such an architecture can provide monotonously decreasing tracking error reductions online. The robustness of the algorithm to changes in the inverse kinematic component and morphology of the soft robot itself is shown.

Dynamic control of soft robotic manipulators is an open problem yet to be well explored and analyzed. Most of the current applications of soft robotic manipulators utilize static or quasi-dynamic controllers based on kinematic models or linearity in the joint space. However, such approaches are not truly exploiting the rich dynamics of a soft-bodied system. In this paper, we present a model-based policy learning algorithm for closed-loop predictive control of a soft robotic manipulator. The forward dynamic model is represented using a recurrent neural network. The closed-loop policy is derived using trajectory optimization and supervised learning. The approach is verified first on a simulated piecewise constant strain model of a cable driven under-actuated soft manipulator. Furthermore, we experimentally demonstrate on a soft pneumatically actuated manipulator how closed-loop control policies can be derived that can accommodate variable frequency control and unmodeled external loads.

Design of continuum manipulators have always been restricted to simplified structures due to the modeling complexity involved. However with the success of machine learning based models, there are potentially more structures and actuation strategies that could be investigated. This paper presents a novel continuum manipulator with variable diameter and a general control strategy for closed loop task space control. The controller constitutes of an inverse kinematics based feedback component, a forward kinematics based feedforward component and a low level velocity controller. Our findings indicate how such an approach can deal with singularities and provide smooth and accurate motion.

With the rise of soft robotics technology and applications, there have been increasing interests in the development of controllers appropriate for their particular design. Being fundamentally different from traditional rigid robots, there is still not a unified framework for the design, analysis, and control of these high-dimensional robots. This review article attempts to provide an insight into various controllers developed for continuum/soft robots as a guideline for future applications in the soft robotics field. A comprehensive assessment of various control strategies and an insight into the future areas of research in this field are presented.

The idea of using embodied intelligence over traditional well-structured design and control formulations has given rise to simple yet elegant applications in the form of soft grippers and compliant locomotion-based robots. Real-world applications of soft manipulators are however limited, largely due to their low accuracy and force transmission. Nonetheless, with the rise of robotic appliances in the field of human-robot interaction, their advantages could outweigh their control deficiencies. In this context, the embodied intelligence could play an important role in developing safe and robust controllers. In this paper, we present a three module soft manipulator to experimentally demonstrate how its morphological properties can be exploited through interactions with the external environment. In particular, we show how to improve the pose accuracy in an assistive task using a simple control algorithm. The soft manipulator takes advantage of its inherent compliance and the physical constraints of the external environment to accomplish a safe interactive task with the user. There exists a continuous and mutual adaptation between the soft-bodied system and the environment. This feature can be used in tasks where the environment is unstructured (e.g. specific body region), and the adaptability of the interaction is entirely dependent on the morphology and control of the system. Experimental results indicate that significant improvements in the tracking accuracy can be achieved by a simple yet appropriate environmental constraint.

Dynamic control of soft robotic manipulators is a challenging field still in its nascent stages. Modeling is still a major hurdle due to its high dimensional nonlinear dynamic properties. Even if accurate models of these high dimensional nonlinear systems are available, the computational burden of the models or sensory requirements poses problems for robust and stable control. This letter addresses these two problems by using a data-driven model that uses only mechanical feedback for stable control of the soft robotic manipulator. Using a learning-based open loop dynamic controller, we investigate the self-stabilizing behavior that can be obtained from the complex dynamics of a soft manipulator. Experimental findings illustrate the advantage of using open-loop dynamic controllers for accurate self-stabilizing control of soft robotic manipulators without any sensory feedback.