It is widely recognized that knowledge sharing and learning are most effective when occurring in an appropriate and supportive context. As a consequence when designing an educational process the context has to be carefully considered. Implementing "situated learning" is straightforward in traditional apprenticeship models that inspired the communities of practice model, and the centuries old university model, but poses a formidable challenge when knowledge and associated practices are to be shared with large numbers of students as it is the case in most contemporary, demand-driven universities and even more in MOOCs. The traditional, and well tested, face-to-face teaching model that is in place does not scale even if videoconferencing is utilized to include students at remote locations. A few years ago, MOOCs were presented as a viable "solution", but while participant numbers remain impressive, experiences also suggest that the learning experience does not scale as easily. In this paper we discuss an alternative knowledge sharing and learning model, the Shangh AI Lectures (SHAIL) developed at the University of Zurich, Switzerland in response to the need of making top quality higher education available to a worldwide audience. SHAIL combines some of the strongest aspects of traditional face-to-face teaching, but also allows lecturing from and to a community which is globally distributed. Findings from delivering the Shangh AI Lectures annually over five consecutive years to up to twenty universities from across the globe (North America, Africa, Europe, Asia, Oceania) suggest that SHAIL is a promising way to address "global teaching", but there are also challenges to be addressed such as: finding a common base with respect to university requirements, student goals and expectations, teaching styles, and of course the actual technologies used.

In recent years, several studies proposed the application of echo state networks (ESN) to adaptive reinforcement learning schemes for the control of artificial autonomous agents. Especially the actor-critic design (ACD) is a promising candidate for robotic systems with continuous state and action spaces, as was demonstrated in several studies using simple wheeled robots. In the present work, we investigate applicability of this learning framework to more complex robotic systems, namely a quadruped running robot with rich dynamics. New challenges and questions arise, such as the nontrivial mapping of actions to the resulting behavior.

Soft robots possess several potential advantages over traditional articulated ones and have attracted significant interest in recent years. However, to control this new type of robots using conventional model-based robotic control approaches is generally ineffective. In this paper, we investigate the challenge to embed and switch among multiple behaviors for an octopus-inspired soft robotic arm. An online learning method for reservoir computing is exploited for this task. This online learning method does not require a separate teaching data collection phase; thus, it has the potential to achieve autonomy in soft robots. Our result shows the feasibility of this approach.

Octopus arms, as well as elephant trunks, squid tentacles, and vertebrate tongues are termed muscular-hydrostats. In such structures, the volume of the organ remains constant during their motions, enabling diverse, complex, and highly controlled movements without the support of a skeleton. Such flexible structures show major advantages over articulated arms that have a rigid skeleton and joints. These advantages have been attracting roboticists aiming to apply these material properties to soft robot controls. In this paper, we show that the muscular-hydrostat system itself has the computational capacity to achieve a complex nonlinear computation. By using a 3D dynamic simulator of the system inspired by the octopus, we actually demonstrate that the system is capable of emulating complex nonlinear dynamical systems by exploiting its elastic body dynamics as a computational resource. In addition, we systematically analyze its computational power in terms of memory capacity, and show that the system has an intrinsic and characteristic short term memory profile. Finally, the implications for soft robot control and future application scenarios are discussed.

Aim at enhancing dexterous and safe operation in unstructured environment, a cable-driven soft robotic manipulator is designed in this paper. Due to soft material it made of and nearly infinite degree of freedom it owns, the soft robotic manipulator has higher security and dexterity than traditional rigid-link manipulator, which make it suitable to perform tasks in complex environments that is narrow, confined and unstructured. Though the soft robotic manipulator possesses advantages above, it is not an easy thing for it to achieve precise position control. In order to solve this problem, a kinematic model based on piecewise constant curvature hypothesis is proposed. Through building up three spaces and two mappings, the relationship between the length variables of 4 cables and the position and orientation of the soft robotic manipulator end-effector is obtained. Afterwards, a depth-independent image Jacobian matrix is introduced and an image-based visual servo controller is presented. Applied by adaptive algorithm, the controller could estimate unknown position of the feature point online, and then Lyapunov theory is used to prove the stability of the proposed controller. At last, experiments are conducted to demonstrate rationality and validity of the kinematic model and adaptive visual servo controller.

Recent results suggest that compliance and non-linearity in physical bodies of soft robots may not be disadvantageous properties with respect to control, but rather of advantage. In the context of morphological computation one could see such complex structures as potential computational resources. In this study, we implement and exploit this view point in a spine-driven quadruped robot called Kitty by using its flexible spine as a computational resource. The spine is an actuated multi-joint structure consisting of a sequence of soft silicone blocks. Its complex dynamics are captured by a set of force sensors and used to construct a closed-loop to drive the motor commands. We use simple static, linear readout weights to combine the sensor values to generate multiple gait patterns (bounding, trotting, turning behavior). In addition, we demonstrate the robustness of the setup by applying strong external perturbations in form of additional loads. The system is able to fully recover to its nominal gait patterns (which are encoded in the linear readout weights) after the perturbation has vanished.

The purposeful action of any agent in a complex environment requires control, that is, the determination of specific sequences of values for the parameters determining the state of the agent. In recent years, it became clear that we have to extend our notion of control if we want to understand the mechanical and chemical process management of biological systems. As it turns out, the lessons we learn from this extension can be directly used in engineering—foremost in the field of robotics, but increasingly also in other areas, such as artificial life or novel types of chemical systems design. The extension we refer to includes the intrinsic dynamics of the system to be controlled as an active, even computational element of control. The employment of the physical or chemical dynamics of a system as part of the computations necessary for control is the underlying principle of the concept of morphological computation. In our use of the term “morphology” we include all aspects of a physical system, not only the shape of the agent, the geometrical combination of the body parts, but also its material properties, such as friction coefficients or parameters describing compliance. Moreover, we also consider the distribution of sensors and actuators as part of the morphology.Conventionally (and with slight oversimplification), control is understood in terms of an agent (typically the term plant is used) and a separate entity, which is the control unit and consists of a conventional digital computer. The state space of the agent is spanned by a number of parameters, which in the case of robots are mechanical in nature, but in the case of chemical cells can be densities, phase parameters, and so on. The controller possesses, via its sensors, an (at least partial) internal representation of the state of the agent at any time. In addition, the controller can change the state of the agent via some actuators. In an ideal situation, the controller exerts complete control over the agent, which means it can determine the agent's position in state space with high accuracy. In such an ideal situation, the observable behavior is the result of a conventional computation based on the internal representation of the agent in the controller. However, there is always some noise present in the dynamics of the agent. Moreover, if the model of the agent gets too complex (high-dimensional state space, high nonlinearity, noise, etc.), the required controller may be intractable. To avoid both problems, that is, to keep the agent controllable, an often followed strategy (during the design process) consists in the reduction of the number of degrees of freedom of the agent. For example, in the case of robots this is reflected by the use of rigid body parts, which are connected by high-torque servos.Control based on morphological computation pursues a different strategy. There is still a division into agent and controller. However, the controller is no longer assumed to exert complete control, and therefore the line of separation gets fuzzy. In extreme cases (e.g., mechanically driven walking machines like passive walkers [1]) this separation can even vanish completely. Morphological computation requires that the physical (chemical) dynamics of the agent be designed in such a way that the complexity of the task that is solved by the controller is minimized. This happens by outsourcing parts of the computation to the physical body (more on the precise meaning of outsourcing later). In a simple example the body dynamics of the agent provide various limit cycles, and the task of the controller is reduced to choosing the right basin of attraction and initiating jumps into it. The choice will clearly depend on the demands made by the environment and/or the prescribed task. For example, in the case of a locomoting robot the dynamics of the body implicitly take over tasks such as the stabilization of the movement pattern against small perturbations, while the controller simply has to choose an appropriate gait. Stated differently, the number of tasks to be fulfilled by the conventional controller is reduced. This reduction can be understood as an outsourcing of computation to the body, since the control problem as such (e.g., how to handle small perturbations caused by an uneven floor in the case of walking machines) is still the same. Notably, not only can a clever exploitation of the physical dynamics of an agent reduce the amount of computation to be performed by the controller, but also the information necessary for the internal representation of the agent's state is reduced. As soon as the stabilization of a movement pattern is intrinsic to the body dynamics of the moving agent, the controller only needs to know in which basin of attraction the agent actually resides. No details about positions, velocities, and so on are necessary. In consequence, the computations that remain to the controller are not necessarily full simulations of the dynamics of the agent, but relatively simple evaluations such as “If condition A is satisfied and agent resides in basin of attraction B, jump into basin C”—a task that is mastered by a simple finite state automaton.It should be emphasized that morphological computation is more than clever systems design. Our claim that computation can be outsourced to the body and not just omitted is corroborated by a recent result of Hauser et al. [3, 4]. They demonstrated that a sufficiently complex physical body equipped with a simple readout (e.g., a feed-forward neural network or even only a linear and static readout) is computationally equivalent to a conventional device almost as powerful as a Turing machine. This finding has two major implications. The first one is that computation can be outsourced to the body in a quite literal sense. The second one is relevant for the engineer's choice of the design of an agent. There is a whole spectrum of options for distributing computations between the physical dynamics of the agent and a conventional control device. It is quite plausible that the optimal choice will not be found at either end of this spectrum. As briefly mentioned in the introduction, the agents under consideration need not be classical, macroscopic robots. It is well known that chemical process management in eukariotic cells results from an interplay between molecular, supramolecular, and mesoscopic structures (e.g., membranes or the cytoskeleton) and their dynamic interaction. For an illustrative example, consider protein sorting by vesicular transport [5].Recently, simple forms of cognition and motility have been implemented in artificial systems in the laboratory; see [2]. Such systems, besides being of fundamental interest for questions concerning the origin of life as well as modern bionanoengineering, also shed light on a form of morphological control not directly visible in the macroscopic world. In the study of their dynamics, classical mechanics is replaced, or at least complemented, by statistical physics. Furthermore, collective phenomena, such as phase transitions, start to play an important role.Morphological computation can be found on a wide range of different scales in a number of different contexts. Consequently, it has attracted many scientists from various backgrounds. For example, roboticists are interested in adopting this concept to build more energy-efficient and robust robots. Biologists try to understand the functions of body parts from the viewpoint of morphology (i.e., morphofunctionality). Computational scientists aim to capture the implications a physical implementation of certain types of computation has for computational theory as a whole. Scientists in the field of medicine have been attracted by the idea of morphological computation. For example, the decrease of movement control in elderly patients can, from the viewpoint of morphological computation, be interpreted as the loss of (morpho)computational power and is not necessarily the consequence of decreasing neural activity. The cause of this loss can be found in a change of the mechanical properties of the body—a change that also alters the attractor landscape in the state space of the body. As a result of aging, this attractor landscape may become less pronounced (e.g., self-stabilization of movement patterns may become slower) and be more difficult to navigate.The interdisciplinarity of the research community suggests that morphological computation is a very general concept. On one hand, this represents a significant challenge when we try to capture the concept in a unified theory or even if we want to define the boundaries of the field itself. On the other hand, it provides us with numerous research opportunities, especially in the interdisciplinary, overlap areas of research. We believe morphological computation even has the potential to give rise to completely new fields of research. To harvest this inherent potential of our community, we organized the 2nd International Conference on Morphological Computation (http://morphcomp.org/) in September, 2011, in Venice. As one of the results of the conference we initiated this special issue. The conference committee and all invited speakers, all experts in the field, were asked to vote for their favorite submissions among the accepted abstracts. Based on this poll, we invited the corresponding authors to submit an extended journal version of their work.As a result, we have in total nine accepted articles, which provide us with a snapshot of recent research done in the context of morphological computation. Furthermore, they also point to new directions of research in the field. While some of the articles have a focus on engineering (e.g., building a series of real robots to investigate the influence of morphology on locomotion, as in Reis et al.'s article), others are oriented toward more general considerations and point to potential future research directions. One from the latter category is the submission by Herrera and Sanz. Their article illustrates impressively how morphological computation attracts an interdisciplinary audience. The background of the authors is found in the theory of emotion. They suggest linking emotions with changes in the morphology and the resulting changed functionality. For example, if the emotion of fear arises, the body changes, namely, it gets ready—either to fight or to flee. As a consequence, the body can no longer be described as a fixed dynamical system, but has to been seen as a variable dynamical structure, which changes its state of attraction (from a finite and discrete set) depending on the emotion it embodies at the moment. This approach implies a new way of looking at the embodiment in robotics and constitutes a new challenge for the design and control of such physical bodies. Although such emotion-induced changes will take place on a slower timescale than the usual movements, the control is nontrivial. The controller will have to produce discrete switches between analogue systems in a stable fashion. The authors suggest some potential control theoretical approaches; however, we might have to rethink the control approach for such systems entirely.Another article, which also belongs to the more general category of this special issue, is the one by Füchslin et al. It consists of two parts: First, it presents and discusses a formalization of the concept of morphological computation. The authors build upon the Venice definition1 given at the First International Conference on Morphological Computation. They clarify the notions of result and input, and the notion and role of a program in morphological computation. Such a formalization is helpful when one aims at a better understanding of what it means to program a morphological computation. It is often illustrated using attractor landscapes of dynamical systems. Füchslin et al. discuss also what it means if the controller (the brain) can alter the attractor landscape. They interpret, for example, training in various sport activities not as learning how to move, but learning how to bring the body into a morphology (internal tension, geometry) that is optimally suited for a given task. Based on different types of feedback between agent and controller (body and brain), they present a hierarchy of levels of morphological computation. In a second part, they present three partially ongoing case studies of systems employing morphological computation and control in different contexts. A first case study deals with exoskeletal support systems for people with movement impairments. In a second investigation, they analyze self-assembly processes from the viewpoint of morphological computation. The third study is more speculative and discusses aspects of cellular dynamics in a specific type of combination therapy in oncology with respect to control.Umedachi et al. investigated morphological computation on the level of single-cell organisms—specifically, the true slime mold. This is a large amoeba-like unicellular organism that does not have any nervous it a of In to understand how such complex patterns can any even any simple neural they implemented a simple physical In their simulations they changes of behavior as and by mechanically the agent, the parameters of the model itself. This clearly to the of the physical which the of complex in that can shed light on how biological systems can based on a clever the underlying will us to build robots with the of and behavior in the physical This is especially for systems, that is, systems no controller is present to what to the concept of morphological computation implies that a computation can be at least in part, by the physical body of an agent, it is to to what this of a computational task can be should and how is therefore for the questions have been investigated by and In their they a simple model of a robot in They to the robot and an optimal controller by learning to They a to capture the of the found and how of the computational task been outsourced to the physical body. They investigated different parameters, like friction and of different body parts, and their influence on the of computation from the controller to the physical part of the They even to the body and for optimal Their results clearly how computation can be to the body, resulting in at a While their of complexity is to their (i.e., to their learning their results point to a research to more general to morphological idea of morphological computation implies that the design process for robots will have to be in robot such as and high are to be for a computationally powerful robot Hauser et al. [3, As a consequence, new types of design and a novel of robots will A in that has been made by et al. in this issue. of using classical for their robots (i.e., they investigated Their choice based on the properties of structures and not made they any specific or they a given task This is a new way of looking at robots by morphological computation. The authors how such physical systems can when only by very simple the complex is done the physical the resulting controller on of it can be simple to the of being linear and As a consequence, the controller parameters can be This to the that morphology not only for but also the corresponding learning task. This might be an of the that biological systems, their complex are and are a to investigate different However, there are as it is not to of the real world. a approach to understand underlying of morphological computation is to build robots with different and investigate For example, Reis et al. a series of robots with specific in their This to investigate the influence of certain morphological in a way and to the results with their They demonstrated that quite a simple can various their case different and how different can be by morphology we not only refer to the but also include material properties, of sensors and and so on. et al. in their on one friction in the case of suggest that friction an important if The authors a simple model and their with on real that friction is in robot as it is as a cause of However, morphological computation an point of and suggests if the design has been made friction can be of et al. investigated morphological computation in the context of Although they not robots on the they investigate how morphology under such no controller, give rise to more complex and Their study deals with in which and a They that the right morphology can give rise to in a robust fashion. Their results may shed light on processes in Their macroscopic system is consists of and may as an for the study of processes as they in cellular process looking at the various of morphological computation that can be found in nature, it is a approach to take a at the of biological systems. One approach is, for example, to build robots and see how this specific morphology is of The this et in this issue, their results of of and also the they have by this the they of an in their complexity and number of They a of the robot and a number of control of the are based on classical control while others use learning to with the complex approach made by using as a control In any such a complex (from the viewpoint of system theory this is a very and is a task. The results are one of the to the for research in the context of an community of scientists morphological computation is a and As this special results have been so but they also point to a number of research is a of research directions for morphological we the concept of morphological computation, we to an new way to at As a consequence, the physical body of a robot will have to be more in the design process than it is We believe the change will be even will not only but will also be of new types of that have properties that support morphological computation. et with their use of structures as a have in that the robots by Reis et al. a new in that they not in form to any to this special issue, by et also us to classical robot design. Their can be found on the and artificial forms of with respect to morphological computation can novel directions. In one of their case Füchslin et al. discuss a support system for patients with movement that aims at of the body, but does not directly this new of robots based on new design will be in general more and of new types of robots is that the clear separation between the controller and the will Such a be by classical control However, even new types of robots will have to be we are to need a new type of control, to classical control which we refer to as The controller can be like the of an The are the body parts, their and They are to a certain and are not controlled at of but they are by the To give an example, in the case of a locomoting the be body parts, that is, mechanical structures with their sensors and which are to small perturbations on a they be as stable limit The only have to between different (i.e., different limit to to to the control problem is the task of learning to control the body. As out, the use of a clever morphology can to reduce the complexity of the corresponding learning task. and in their an approach to a simple model of a However, robots that will take of morphological computation will be very and We will have to new to understand how learning can take place The learning will have to take of the and should be to the given is to the controller with the body. While this has to be done in simulations for we in the future to see of and to see in we into the physical body, we on one in that we can computation to the but on the other we There is a space we need to understand to morphological computation in that context are passive walkers all their computation place in the physical (i.e., in the mechanical of the robot and its with the While it is to see such robots completely when walking under different We also have to consider that the physical properties of the material used For example, there is a for to they as Furthermore, we will have to the of mechanical structures when they are for research is to theoretical for morphological computation. It is to have to which properties are for computation. It has been only that there has been some in this For example, it has been by Hauser et al. [3, by two different for morphological computation, that using physical can to computation is by the concept of a Turing machine. Such a is in morphological computation. Füchslin et in this special issue, a to define morphological computation. As they their relevant aspects of what one as for morphological computation, but there are processes that can be as of morphological computation but are not by their It may be necessary to define types of morphological computation in to keep computation be more the of physical processes by conventional is based on a of the physical into sequences of Füchslin et al. claim that this a of but also a such as the of extended to be difficult to capture by if one does not believe in a fundamental between conventional and morphological computation the a also is a physical theoretical considerations should clarify there are systems that one can as and and that that are computationally when using conventional connected to the theoretical is the of and were to define a of complexity for their that directly to their However, it is to more general to the complexity of computation and the One way is to from information However, we believe that new types of might even be in and article, can be for certain However, a real which has to in such a dynamic and complex as has to be to with a number of different is a for robots. This means that the morphology has to be changed to the task. We believe that real (and therefore change of some parameters or of and should include is not to provide that robots to it is only a of we have such and we believe morphological computation will from computation has a way from the first of the concept in to we are there is still to The boundaries of the field are not well a is and are However, we believe we are right in an for morphological computation as one can see in this special issue. The community of roboticists to be the concept. There is a clear in building especially with to with is in new to morphological computation. the research community morphological computation is We are looking to more results and a new of robots. the idea of morphological computation will be into a by and like to all the of this special for their and their helpful We also like to and for their support and their

Self-organization is a phenomenon found in biomolecular self-assembly by which proteins are spontaneously driven to assemble and attain various functionalities. This study reports on self-organized behavior in which distributed centimeter-sized modules stochastically aggregate and exhibit a translational wheeling motion. The system consists of two types of centimeter-sized water-floating modules: a triangular-shaped module that is equipped with a vibration motor and a permanent magnet (termed the active module), which can quasi-randomly rove around; and circular modules that are equipped with permanent magnets (termed passive modules). In its quasi-random movement in water, the active module picks up passive modules through magnetic attraction. The contacts between the modules induce a torque transfer from the active module to the passive modules. This results in rotational motion of the passive modules. As a consequence of the shape difference between the triangular module and the circular module, the passive modules rotate like wheels, being kept on the same edges as the active module. The motion of the active module is examined, as well as the characteristics and behavior of the self-organization process.

Anthropomimetic robotics differs from conventional approaches by capitalizing on the replication of the inner structures of the human body, such as muscles, tendons, bones, and joints. Here we present our results of more than three years of research in constructing, simulating, and, most importantly, controlling anthropomimetic robots. We manufactured four physical torsos, each more complex than its predecessor, and developed the tools required to simulate their behavior. Furthermore, six different control approaches, inspired by classical control theory, machine learning, and neuroscience, were developed and evaluated via these simulations or in small-scale setups. While the obtained results are encouraging, we are aware that we have barely exploited the potential of the anthropomimetic design so far. But, with the tools developed, we are confident that this novel approach will contribute to our understanding of morphological computation and human motor control in the future.

Stable grasp is the result of sensorimotor regulation of forces, ensuring sufficient grip force and the integrity of the held object. Grasping with a prosthesis introduces the challenge of finding the appropriate forces given the engineered sensorimotor prosthetic interface. Excessive force leads to unnecessary energy use and possible damage to the object. In contrast, low grip forces lead to slippage. In order for a prosthetic hand to achieve a stable grasp, the haptic information provided to the prosthesis wearer needs to display these two antagonistic grasp metrics (force and slip) in a quantified way. We present the design and evaluation of a wearable single-actuator haptic device that relays multi-modal haptic information, such as grip force and slip speed. Two belts that are activated in a mutually exclusive manner by the rotation direction of a single motor exert normal force and tangential motion on the skin surface, respectively. The wearable haptic device is able to display normal forces as a tap frequency in the range of approximately 1.5–5.0 Hz and slip speed in the range of 50–200 mm/s. Within these values, users are able to identify at least four stimulation levels for each feedback modality, with short-term training.

Soft robots have significant advantages over traditional robots made of rigid materials. However, controlling this type of robot by conventional approaches is difficult. Reservoir computing has been demonstrated to be an effective approach for achieving rapid learning in benchmark tasks and conventional robots. In this study, we investigated the feasibility and capacity of the reservoir computing approach to embedding and switching between multiple behaviors in a on-line manner in a soft robotic arm. The result shows that this approach can successfully achieve this task.

Soft robots are difficult to control because of their compliant and elastic body dynamics compared with robots made of rigid bodies. In this paper, we present a control scheme inspired by the octopus called timing-based control for soft robotic arms. This control scheme is motivated to positively exploit the natural dynamics of the soft body. We demonstrate a scheme for controlling an object-reaching task by using an echo state network on a 3D physical soft robotic arm simulator and show that this network can successfully perform the task. Detailed analyses and evaluations of the generalization capacity of the network and the performances to the reaching task are presented.

The biological hypothesis of spinal engine states that locomotion is mainly achieved by the spine, while the legs may serve as assistance. Inspired by this hypothesis, a compliant, multiple degree-of-freedom, biologically-inspired spine has been embedded into a quadruped robot, named Kitty, which has no actuation on the legs. In this paper, we demonstrate how versatile behaviors (bounding, trotting, and turning) can be generated exclusively by the spine's movements through dynamical interaction between the controller, the body, and the environment, known as embodiment. Moreover, we introduce information theoretic approach to quantitatively study the spine internal dynamics and its effect on the bounding gait based on three spinal morphologies. These three morphologies differ in the position of virtual spinal joint where the spine is easier to get bent. The experimental results reveal that locomotion can be enhanced by using the spine featuring a rear virtual spinal joint, which offers more freedom for the rear legs to move forward. In addition, the information theoretic analysis shows that, according to the morphological differences of the spine, the information structure changes. The relationship between the observed behavior of the robot and the corresponding information structure is discussed in detail.

Compliant actuation contributes enormously in legged locomotion robotics since it is able to alleviate control efforts in improving the robot's adaptability and energy efficiency. In this paper, we present a novel design of a variable stiffness rotary actuator, called MESTRAN, which was especially targeted to address the limitations in terms of the amount of energy and time required to vary the stiffness of an actuated joint. We have constructed a mechanical model in simulation and a physical prototype. We conducted a series of experiments to validate the performance of the MESTRAN actuator prototype. The results from the simulation and experiments show that MESTRAN allows independent control of stiffness and position of an actuated rotary joint with a large operational range and high speed. The torque-displacement relationship is close to linear. Lastly, the MESTRAN actuator is energy-efficient since a certain stiffness level is maintained without energy input.

Fish excel in their swimming capabilities. These result from a dynamic interplay of actuation, passive properties of fish body, and interaction with the surrounding fluid. In particular, fish are able to exploit wakes that are generated by objects in flowing water. A powerful demonstration that this is largely due to passive body properties are studies on dead trout. Inspired by that, we developed a multi joint swimming platform that explores the potential of a passive dynamic mechanism. The platform has one actuated joint only, followed by three passive joints whose stiffness can be changed online, individually, and can be set to an almost arbitrary nonlinear stiffness profile. In a set of experiments, using online optimization, we investigated how the platform can discover optimal stiffness distribution along its body in response to different frequency and amplitude of actuation. We show that a heterogeneous stiffness distribution each joint having a different value outperforms a homogeneous one in producing thrust. Furthermore, different gaits emerged in different settings of the actuated joint. This work illustrates the potential of online adaption of passive body properties, leading to optimized swimming, especially in an unsteady environment.

We introduce a simple model of human's musculoskeletal system to identify the computation that a compliant physical body can achieve. A one-joint system driven by actuation of the springs around the joint is used as a computational device to compute the temporal integration and nonlinear combination of an input signal. Only a linear and static readout unit is needed to extract the output of the computation. The results of computer simulations indicate that the network of mechanically coupled springs can emulate several nonlinear combinations which need temporal integration. The simulation with a two-joint system also shows that, thanks to mechanical connection between the joints, a distant part of a compliant body can serve as a computational device driven by the indirect input. Finally, computational capability of antagonistic muscles and information transfer through mechanical couplings are discussed.

In this paper, we investigate possible means of communication between plants and machines. Plants are capable of sensing a variety of environmental information. In particular, Avocado trees have an apparent response to increasing drought levels. We read out two different communication channels: morphological changes (leaf inclination) and the electric potential of the stem (biopotential) using distance sensors and biopotential electrodes, respectively. Leaf inclination reliably triggers irrigation, whereas the changes of the biopotential indicate water uptake and can be used to automatically stop the irrigation. Hence, through systematic experiments we demonstrate that morphological changes and biopotentials provide suitable control signals for interfacing plants with machines, and open a possibility to exploit abilities of plants in robotic systems.

The human hand is one of the most complex structures in the body, being involved in dexterous manipulation and fine sensing. Traditional engineering approaches have mostly attempted to match such complexity in robotics without sufficiently stressing on the underlying mechanisms that its morphology encodes. In this work, we propose an artificial skin able to encode, through its morphology, the tactile sense of a robotic hand, characteristic to slippage events. The underlying layout consists of ridges and allows slippage detection and the quantification of slippage speed. Such encoding of slippage signal becomes suitable for relaying tactile feedback to users in prosthetic applications. This approach emphasizes the importance of exploiting morphology and mechanics in structures for the design of prosthetic interfaces.

Self-assembly is a process in which individual components form an organized structure as a consequence of local interactions. When using magnetics to create interaction forces, the magnetic flux distribution of a self-assembling system changes as its assembly state varies. Since Hall effect sensors are a convenient and effective means to detect changes in the magnetic field, we explore their applicability to monitoring the morphology of such magnetically self-assembling systems. We find that optimal positions for the sensor can be found where the flux changes maximally. Our analysis is applied to two different systems by deriving the flux changes for all possible states, and theoretical flux changes are verified with experiments. In addition, we show that a small number of sensors is sufficient for robust state determination. In addition to state detection, experiments show the potential for angle measurement for compliant cylindrical magnet joints using a single Hall sensor.

This work presents a multimodal bottom-up attention system for the humanoid robot iCub where the robot's decisions to move eyes and neck are based on visual and acoustic saliency maps. We introduce a modular and distributed software architecture which is capable of fusing visual and acoustic saliency maps into one egocentric frame of reference. This system endows the iCub with an emergent exploratory behavior reacting to combined visual and auditory saliency. The developed software modules provide a flexible foundation for the open iCub platform and for further experiments and developments, including higher levels of attention and representation of the peripersonal space.

We present a novel type of inter-module connection mechanism for waterborne modular robotic systems. The proposed mechanism exploits the thermoelectric effect to cool down and freeze the water between two modules thus causes them to attach to each other. We validate the feasibility of this mechanism by embedding a Peltier heat pump (m = 0.8 g) in two types of cm scale self-assembly systems, one in which the modules are free to move and one in which the modules are linked together by hinges. Our experimental results demonstrate that the proposed Peltier-based connector has (a) a high bond strength/weight ratio for a rather large range of temperatures and (b) is rather robust against misalignments between docking modules, making it a useful alternative to current connection mechanisms for small scale low autonomy self-assembly systems.

We present a novel type of inter-module connection mechanism for waterborne modular robotic systems. The proposed mechanism exploits the thermoelectric effect to cool down and freeze the water between two modules thus causes them to attach to each other. We validate the feasibility of this mechanism by embedding Peltier heat pumps (m = 0.8 g) in a cm scale self-assembly system. Our experimental results demonstrate that the proposed Peltier-based connector has (a) a high bond strength/weight ratio for a rather large range of temperatures and (b) is rather robust against misalignments between docking modules, making it a useful alternative to current connection mechanisms for small scale low autonomy self-assembly systems.

When building autonomous robotic agents, designers almost always make assumptions about how the control system relates sensory information to motor actions, in order to provide the agent with a set of basic behaviors. This raises the question of how arbitrary these assumptions are, and to what extent the introduced biases reduce the potential for the generation of new behaviors arising from interaction with the environment. In this paper, we propose a new model of robot control architecture, consisting merely of homogeneous, non-hierarchical sensorimotor coupling. We show that a robot using this model can display coherent complex behaviors which emerge from the agent-environment interaction, such as tracking an object, or solving a task based on the temporal relationship between an early clue and a delayed reward.

New robotics is an approach to robotics that, in contrast to traditional robotics, employs ideas and principles from biology. While in the traditional approach there are generally accepted methods (e. g., from control theory), designing agents in the new robotics approach is still largely considered an art. In recent years, we have been developing a set of heuristics, or design principles, that on the one hand capture theoretical insights about intelligent (adaptive) behavior, and on the other provide guidance in actually designing and building systems. In this article we provide an overview of all the principles but focus on the principles of ecological balance, which concerns the relation between environment, morphology, materials, and control, and sensory-motor coordination, which concerns self-generated sensory stimulation as the agent interacts with the environment and which is a key to the development of high-level intelligence. As we argue, artificial evolution together with morphogenesis is not only "nice to have" but is in fact a necessary tool for designing embodied agents.

The goal of this paper is to explore whisker systems as a sensory modality in animals (rodents) and robots with respect to their adaptive potential. To this end, we have developed an active whisker system. Whiskers from real mice, when attached to capacitor microphones, can detect very weak stimuli at their tips. Different eigen frequencies are induced in the whisker according to the materials over which the whiskers move. Whisking is active, i.e., they can move back and forth to induce sensory stimulation even when an agent is not moving. In this study, active whisking is applied to navigation tasks. The results show that materials can be distinguished using 2-3 kHz signals, which correspond to the experimental evidence obtained from animals.

Whiskers are powerful sensors for robots that are not only useful for basic tasks such as obstacle avoidance, but also have the potential for gathering rich information about objects. We have developed an active multi-whisker array modeled on the rat whisker system which can be amounted on a mobile robot. We show that with this whisker array we can discriminate different textures based on the frequencies elicited by the whiskers. We exploit the phase-locked structure of our data using sensory-motor integration. Two factors enable better discrimination of the textures: firstly, considering several touch events from one whisker; and secondly, combining the information from more than one whisker.

Developmental robotics is an emerging field located at the intersection of robotics, cognitive science and developmental sciences. This paper elucidates the main reasons and key motivations behind the convergence of fields with seemingly disparate interests, and shows why developmental robotics might prove to be beneficial for all fields involved. The methodology advocated is synthetic and two-pronged: on the one hand, it employs robots to instantiate models originating from developmental sciences; on the other hand, it aims to develop better robotic systems by exploiting insights gained from studies on ontogenetic development. This paper gives a survey of the relevant research issues and points to some future research directions.

In this paper, we present a first series of experiments with prototype artificial whiskers that have been developed in our laboratory. These experiments have been inspired by neuroscience research on real rats. In spite of the enormous potential of whiskers, they have to date not been systematically investigated and exploited by roboticists. Although the transduction mechanism is simple and straightforward, and the whiskers are currently used in a passive way only, the dynamics of the sensory signals resulting from the interaction with various textured surfaces is complex and has a rich information content. The experiments provide the foundation for future work including active sensing, whisker arrays, and cross-modal integration.