The increasing demand for physical interaction between humans and robots has led to an interest in robots that guarantee safe behavior when human contact occurs. However, attaining established levels of performance while ensuring safety creates formidable challenges in mechanical design, actuation, sensing and control. To promote safety without compromising performance, a human-friendly robotic arm has been developed using the concept of hybrid actuation. The new design employs high-power, low-impedance pneumatic artificial muscles augmented with small electrical actuators, distributed compact pressure regulators with proportional valves, and hollow plastic links. The experimental results show that significant performance improvement can be achieved with hybrid actuation over a system with pneumatic muscles alone. In this paper we evaluate the safety of the new robot arm through experiments and simulation, demonstrating that its inertia/power characteristics surpass those of previous human-friendly robots we have developed.

Robotic cell microinjection is a technique that utilizes automation technology to insert substances into a single living cell with a fine needle. Compared with manual microinjection, the main benefits of the robotic cell injection are quality, productivity and repeatability. In this paper we aim to control the penetration force during robotic cell injection to quantify the influence of the penetration force on cells. A force-control-based cell injection approach that is capable of regulating the penetration force in a desired force trajectory is developed. The proposed force control framework includes two control loops. The inner loop is an impedance control used to specify the interaction between the needle and the cell. The outer loop is a force tracking non-linear controller using a feedback linearization technique. The cell model is identified online with a least-squares parameter estimator. With the proposed force control approach, the penetration force can be regulated explicitly to follow the desired force trajectory during the cell injection process. Experiments performed on fish embryos verify the effectiveness of the proposed approach.

We present a complete vision-guided robot system for model-based three-dimensional (3D) pose estimation and picking of singulated 3D objects. Our system employs a novel vision sensor consisting of a video camera surrounded by eight flashes (light emitting diodes). By capturing images under different flashes and observing the shadows, depth edges or silhouettes in the scene are obtained. The silhouettes are segmented into different objects and each silhouette is matched across a database of object silhouettes in different poses to find the coarse 3D pose. The database is pre-computed using a computer-aided design (CAD) model of the object. The pose is refined using a fully projective formulation of Lowe's model-based pose estimation algorithm. The estimated pose is transferred to a robot coordinate system utilizing the hand–eye and camera calibration parameters, which allows the robot to pick the object. Our system outperforms conventional systems using two-dimensional sensors with intensity-based features as well as 3D sensors. We handle complex ambient illumination conditions, challenging specular backgrounds, diffuse as well as specular objects, and texture-less objects, on which traditional systems usually fail. Our vision sensor is capable of computing depth edges in real time and is low cost. Our approach is simple and fast for practical implementation. We present real experimental results using our custom designed sensor mounted on a robot arm to demonstrate the effectiveness of our technique.

Primary challenges in the building of untethered submillimeter sized robots include propulsion methods, power supply, and control. We present a novel type of microrobot called MagMite that utilizes a new class of wireless resonant magnetic micro-actuator that accomplishes all three tasks. The term MagMite is derived from Magnetic Mite—a tribute to the underlying magnetic propulsion principle and the micro-scale dimensions of the robot. The device harvests magnetic energy from the environment and effectively transforms it into inertia- and impact-driven mechanical force while being fully controllable. It can be powered and controlled with oscillating fields in the kilohertz range and strengths as low as 2 mT, which is only roughly 50 times the average Earth magnetic field. These microrobotic agents with dimensions less than 300 µm x 300 µm x 70 µm and a total mass of 30–50 µg are capable of moving forward, backward and turning in place while reaching controllable speeds in excess of 12.5 mm s^–1 or 42 times the robot's body length per second. The robots produce enough force to push micro-objects of similar sizes and can be visually servoed through a maze in a fully automated fashion. The prototype devices exhibit an overall degree of flexibility, controllability, and performance unmatched by other microrobots reported in the literature. The robustness of the MagMites leads to high experimental repeatability, which in turn enabled us to successfully compete in the RoboCup 2007 and 2009 Nanogram competitions. In this work it is demonstrated how the robots exhibit a plethora of driving behaviors, how they can operate on a host of unstructured surfaces under both dry and wet conditions, and how they can accomplish fully automated micromanipulation tasks. Various micro-objects ranging from beads to biological entities have been successfully manipulated. To the same end, multi-agent studies have shown great promise to be used in cooperative tasks.

The problem of generating uniform deterministic samples over the rotation group, SO(3), is fundamental to computational biology, chemistry, physics, and numerous branches of computer science. We present the best-known method to date for constructing incremental, deterministic grids on SO(3); it provides: (1) the lowest metric distortion for grid neighbor edges, (2) optimal dispersion-reduction with each additional sample, (3) explicit neighborhood structure, and (4) equivolumetric partition of SO(3) by the grid cells. We also demonstrate the use of the sequence on motion planning problems.

In this paper we consider the problem of planning sensor observations for a network of overhead sensors which will resolve ambiguities in the output of a horizontal sensor network. More specifically, we address the problem of counting the number of objects detected by the horizontal sensor network, using the overhead network to aim at specific areas to improve the count. We consider several different overhead sensor models. The main theme of our results is that, even though observation planning is intractable for such a network, a simple, greedy algorithm for controlling the overhead sensors guarantees performance with bounded and reasonable suboptimality. Our results are very general and make few assumptions about the specific sensors used.

External actuation in self-reconfigurable modular robots promises to allow modules to shrink down in size. Synchronous external motions promise to allow fast convergence and assembly times. XBot is a modular system that uses synchronous external actuation, but has a limited range of reachable configurations stemming from a single motion primitive of a module rotating about another. This paper proposes to extend the motion primitives by using moves with two modules swinging in a dynamic chain. The feasibility of these motion primitives is proven experimentally. A parameterization of the external actuation motion profiles is explored to define a space of physically valid motion profiles. The larger the space, the more robust the motion primitives will be to inexact initial conditions and to imprecision in the external actuation mechanisms. In addition, this paper proves that a configuration of XBot meta-modules can reach any configuration using just these motion primitives.

The presented multi-user telepresence and teleaction system enables two teleoperators, which are independently controlled by two human operators, to perform collaborative actions in a remote environment. The system provides the operators with visual, auditory, and haptic feedback and allows them to physically interact with objects in the remote environment. The interaction between the two human operators is enhanced by augmenting visual and auditory feedback. The paper is focused on the amalgamation of the individual subsystems, which handle the different modalities, into one tightly integrated system, which creates a common workspace for two operators. The overall system architecture and the appropriate design of the auditory, visual, and haptic subsystems are discussed. The audio system makes use of a novel high-fidelity interpolation technique to render three-dimensional sound scenes for both human operators, which enhances their interaction. The design of the video system allows modeling and rendering of the remote environment in real time while regarding changes in the scene by perpetually updating the model. The haptic system is based on admittance-type devices, which are best fitted for applications involving large workspaces and high interaction forces. In addition, the implementation of locomotion techniques for telepresence in large-scale environments are presented. Finally, an application example shows that the system can be successfully employed in a remote maintenance task, which consists of exploring a large-scale environment, moving to the target area, and finally repairing a broken pipe by attaching a sealing clamp. The example demonstrates the necessity of multi-user telepresence and teleaction systems and supports the benefits of consistent multi-modal feedback.

Steerable needles can be used in medical applications to reach targets behind sensitive or impenetrable areas. The kinematics of a steerable needle are non-holonomic and, in two dimensions, equivalent to a Dubins car with constant radius of curvature. In three dimensions, the needle can be interpreted as an airplane with constant speed and pitch rate, zero yaw, and controllable roll angle. We present a constant-time motion planning algorithm for steerable needles based on explicit geometric inverse kinematics similar to the classic Paden–Kahan subproblems. Reachability and path competitivity are analyzed using analytic comparisons with shortest path solutions for the Dubins car (for two dimensions) and numerical simulations (for three dimensions). We also present an algorithm for local path adaptation using null-space results from redundant manipulator theory. Finally, we discuss several ways to use and extend the inverse kinematics solution to generate needle paths that avoid obstacles.

Motion planning problems encountered in manipulation and legged locomotion have a distinctive multi-modal structure, where the space of feasible configurations consists of intersecting submanifolds, often of different dimensionalities. Such a feasible space does not possess expansiveness, a property that characterizes whether planning queries can be solved efficiently with traditional probabilistic roadmap (PRM) planners. In this paper we present a new PRM-based multi-modal planning algorithm for problems where the number of intersecting manifolds is finite. We also analyze the completeness properties of this algorithm. More specifically, we show that the algorithm converges quickly when each submanifold is individually expansive and establish a bound on the expected running time in that case. We also present an incremental variant of the algorithm that has the same convergence properties, but works better for problems with a large number of submanifolds by considering subsets of submanifolds likely to contain a solution path. These algorithms are demonstrated in geometric examples and in a legged locomotion planner.

We propose a newly renovated mobile robot PEOPLER-II (Perpendicularly Oriented Planetary Legged Robot), and addresses its motion analysis for switching its locomotion from leg-type to wheel-type and vice versa. For the leg-type locomotion, particularly in a transitional state of sitting or standing, we propose a control method based on minimization of the total energy cost using the distribution of the motor power payload in the hip and knee joints, in addition to the method of keeping the same payload factor. Also, we discuss robot configurations for switching between the two locomotion types by considering environmental factors such as walking gaits, ground inclination angle and robot's traveling direction. Knee joint position of a pivotal foot determines knee ahead and knee behind gaits. In each switch, we check such characteristics as the hip joint rotation direction, robot center trajectory, and necessary total power in a practical point of use. Then we build three beneficial switching cycles aiming for moderate use of a motor, rider's comfort, and power saving. Finally, we demonstrate the switching by considering the aim and verify that the results of the analysis become useful for enabling switching on demand.

The solution to the problem of mapping an environment and at the same time using this map to localize (the simultaneous localization and mapping, SLAM, problem) is a key prerequisite in the synthesis of truly autonomous vehicles. By far the most common formulation of the SLAM problem is founded on a vector based stochastic framework, where the sensor models and the vehicle models are represented in state-space form and the joint posterior or its statistics are obtained based on recursive Bayesian estimation. All of these SLAM solutions leading from the stochastic vector state-space approach require that we solve certain parallel problems in each recursion. These include effective solutions to the problems of data association, feature extraction, clutter filtering, and landmark or map management. In this paper, we propose an alternative framework based on finite set statistics (FISST), where the SLAM problem is reformulated so that the landmark map and the measurements are represented using random finite sets and the landmark map is jointly estimated with the vehicle state vector, whilst explicitly accounting for measurement detection uncertainty, data-association uncertainty, false alarms and map management in the SLAM filter framework. Similar to FastSLAM, the proposed formulation is based on a factorization of the full SLAM posterior into a product of the vehicle trajectory posterior and the landmark map posterior conditioned on the vehicle trajectory. The vehicle trajectory posterior is then estimated using a particle filter and the map posterior conditioned on the vehicle trajectory via a sequential Monte Carlo (SMC) implementation of the probability hypothesis density (PHD) filter. Simulation results of the proposed algorithm are presented and benchmarked against FastSLAM to demonstrate the effectiveness and improved performance of the FISST-SLAM in the presence of significant clutter.

The MAV '08 competition focused on the problem of using air and ground vehicles to locate and rescue hostages being held in a remote building. To execute this mission, a number of technical challenges were addressed, including designing the micro air vehicle (MAV), using the MAV to geo-locate ground targets, and planning the motion of ground vehicles to reach the hostage location without detection. In this paper, we describe the complete system designed for the MAV '08 competition, and present our solutions to three technical challenges that were addressed within this system. First, we summarize the design of our MAV, focusing on the navigation and sensing payload. Second, we describe the vision and state estimation algorithms used to track ground features, including stationary obstacles and moving adversaries, from a sequence of images collected by the MAV. Third, we describe the planning algorithm used to generate motion plans for the ground vehicles to approach the hostage building undetected by adversaries; these adversaries are tracked by the MAV from the air. We examine different variants of a search algorithm and describe their performance under different conditions. Finally, we provide results of our system's performance during the mission execution.

In this paper we present a formal approach to robot motion specification. This motion specification takes into account three elementary behaviors that suffice to define any robot interaction with the environment, i.e. free motion, exerting generalized forces and the transition between both of these behaviors. These behaviors provide a foundation for general motion generation taking into account any sensors, any effectors and the capability to exchange information between embodied agents. This specification can be used both for the definition of robot tasks and implementation of robot control software, hence both of those aspects are presented in this paper. This formal approach was used for the implementation of the MRROC++ robot programming framework. Two-handed manipulation of a Rubik's cube is used as an exemplary task. Extensive experimentation both with the presented formalism and the MRROC++ framework showed that the imposed formal rigor eliminates many errors at the software specification phase, produces well-structured control software and significantly speeds up and simplifies its implementation. These advantages are mainly due to the fact that the proposed formal specification tool is derived from operational semantics used in computer science for the definition of programming languages, thus a close relationship between abstract definition and the implementation of the control system resulted.

Quasistatic legged locomotion over uneven terrains requires characterization of the mechanism's feasible equilibrium postures. This paper characterizes the feasible equilibrium postures of mechanisms supported by three frictional point contacts in a three-dimensional gravitational environment, for a subclass of contact arrangements, called tame, for which the friction cones lie above the plane spanned by the contacts. The kinematic structure of the mechanism is lumped into a single rigid body B having the same contacts with the environment and a variable center of mass. The equilibrium postures associated with a given set of contacts become the center-of-mass locations of B that maintain a feasible equilibrium with respect to gravity. The paper establishes the relations between the feasible equilibrium region and the classical support polygon principle. For tame 3-contact arrangements, the paper identifies and characterizes geometrically three types of boundary curves of the feasible equilibrium region, where two of them are obtained is closed-from, and the third is given implicitly as a solution of a set of nonlinear equations, which can be traced numerically. The three types of boundary curves are then associated with the onset of three different modes of non-static contact motions. Finally, the paper reports on experimental results that verify the theoretical predictions by using a 3-legged prototype.

In this paper, we describe the control, navigation, and mapping methods that were developed for a hovering autonomous underwater vehicle that explored flooded cenotes in Mexico. The cenotes of Sistema Zacatón in Tamaulipas, Mexico are flooded sinkholes, exotic geological formations with unique water chemistry. The largest, Zacatón, is over 300-m deep. None of the cenotes were mapped before the present DEPTHX project. The goals of the DEPTHX project were to construct metrically accurate three-dimensional maps of the cenotes, and to collect environmental data, imagery, water samples, and core samples. The unknown extent of the cenotes, together with the challenging scientific mission, spurred the development of a robotic vehicle that autonomously, with no communications to the surface, built accurate three-dimensional maps using sonar and collected a variety of scientific data, including core samples from the cenote walls. In this paper, we describe the design, implementation, and testing of the robot software, as well as the results from mapping four cenotes of Sistema Zacatón.

A lack of sufficient actuation power as well as the presence of passive degrees of freedom are often serious constraints for feasible motions of a robot. Installing passive elastic mechanisms in parallel with the original actuators is one of a few alternatives that allows for large modifications of the range of external forces or torques that can be applied to the mechanical system. If some motions are planned that require a nominal control input above the actuator limitations, then we can search for auxiliary spring-like mechanisms complementing the control scheme in order to overcome the constraints. The intuitive idea of parallel elastic actuation is that spring-like elements generate most of the nominal torque required along a desired trajectory, so the control efforts of the original actuators can be mainly spent in stabilizing the motion. Such attractive arguments are, however, challenging for robots with non-feedback linearizable non-minimum phase dynamics that have one or several passive degrees of freedom. We suggest an approach to resolve the apparent difficulties and illustrate the method with an example of an underactuated planar double pendulum. The results are tested both in simulations and through experimental studies.

The ability to autonomously acquire new knowledge through interaction with the environment is an important research topic in the field of robotics. The knowledge can only be acquired if suitable perception–action capabilities are present: a robotic system has to be able to detect, attend to and manipulate objects in its surrounding. In this paper, we present the results of our long-term work in the area of vision-based sensing and control. The work on finding, attending, recognizing and manipulating objects in domestic environments is studied. We present a stereo-based vision system framework where aspects of top-down and bottom-up attention as well as foveated attention are put into focus and demonstrate how the system can be utilized for robotic object grasping.

Computer simulations play an important role in the design and validation of constrained robotic operations. The fidelity of these simulations, however, depends on the specification of contact dynamics parameters, which often need to be determined experimentally. In this paper we investigate the identification of contact parameters from complex stiff multi-point contact scenarios encountered in typical robotic operations using a recently developed least-squares-based method. This method is extended to deal with geometric uncertainties by employing a non-linear separable least-squares formulation. The latter is solved using a variable projection method, and allows simultaneous identification of contact parameters and geometric uncertainties. The conditions for observability of geometric uncertainties are derived and a regularized formulation is proposed in case the identification of geometric uncertainties is ill-conditioned. The applicability of the original method and the benefits of the extended method with identification of geometric uncertainties for the identification of the contact parameters are illustrated by means of experimental data measured with the Special Purpose Dexterous Manipulator (SPDM) Task Verification Facility (STVF) manipulator at the Canadian Space Agency (CSA).

In this paper we describe a tactile transducer device that is optimized from biomechanical data and that has a compact, yet modular design. The tactile transducer comprises a 6 x 10 piezoelectric bimorph actuator array with a spatial resolution of 1.8 mm x 1.2 mm and has a wide temporal bandwidth. The actuator mounting method was improved from a conventional cantilever method to a dual-pinned method, giving the actuator the ability to deform the glabrous skin maximally during laterotactile stimulation. The results were validated by asking subjects to detect tactile features under a wide range of operating conditions. The tactile display device is modular, makes use of ordinary fabrication methods, and can be assembled and dismantled in a short time for debugging and maintenance. It weighs 60 g, it is self-contained in a 150 cm³ volume and may be interfaced to most computers, provided that two analog outputs and six digital I/O lines are available. Psychophysical experiments were carried out to assess its effectiveness in rendering virtual tactile features.

We present a particle filtering algorithm for visual tracking, in which the state equations for the object motion evolve on the two-dimensional affine group. We first formulate, in a coordinate-invariant and geometrically meaningful way, particle filtering on the affine group that allows for combined state–covariance estimation. Measurement likelihoods are also calculated from the image covariance descriptors using incremental principal geodesic analysis, a generalization of principal component analysis to curved spaces. Comparative visual tracking studies demonstrate the increased robustness of our tracking algorithm.

In this paper, we examine non-stretchable two-dimensional polygonal cloth, and place bounds on the number of fingers needed to immobilize it. For any non-stretchable cloth polygon, it is always necessary to pin all of the convex vertices. We show that for some shapes, more fingers are necessary. No more than one-third of the concave vertices need to be pinned for simple polygons, and no more than one-third of the concave vertices plus two fingers per hole are necessary for polygons with holes.

We introduce a detailed numerical simulation and analysis framework to extend the principles of passive dynamic walking to quadrupedal locomotion. Non-linear limit cycle methods are used to identify possible gaits and to analyze the stability and efficiency of quadrupedal passive dynamic walking. In doing so, special attention is paid to issues that are inherent to quadrupedal locomotion, such as the occurrence of simultaneous contact collisions and the implications of the phase difference between front and back leg pairs. Limit cycles identified within this framework correspond to periodic gaits and can be placed into two categories: in-phase gaits in which front and back legs hit the ground at roughly the same time, and out-of-phase gaits with a ±90° phase shift between the back and front leg pairs. The latter are, in comparison, energetically more efficient but exhibit one unstable eigenvalue that leads to a phase divergence and results in a gait-transition to a less efficient in-phase gait. A detailed analysis examines the influence of various parameters on stability and locomotion speed, with the ultimate goal of determining a stable solution for the energy-efficient, out-of-phase gait. This was achieved through the use of a wobbling mass, i.e. an additional mass that is elastically attached to the main body of the quadruped. The methods, results, and gaits presented in this paper additionally provide a point of departure for the exploration of the considerably richer range of quadrupedal locomotion found in nature.

This paper presents a kinematic motion control strategy for an n-axle Compliant Framed Modular wheeled Mobile Robot (CFMMR). This robot is essentially a passive-joint active-wheel snake robot where coordinated motion of the robot modules is critical for maximizing mobility and minimizing traction forces. A distributed master–slave kinematic motion control structure is proposed where the front axle module of the robot is the master and subsequent axle modules are slaves. An existing path manifold based controller is used to guide the motion of the master. Two steering algorithms with different specializations are then proposed for the slave modules. Performance of the steering algorithms is characterized based upon their capability to reduce traction forces, control final robot posture, and maneuver in a limited space. It is shown that these algorithms satisfy the physical constraints of the robot, which are characterized by path curvature and velocity limitations. Simulation and experimental results validate and characterize the performance of the algorithms.

This paper proposes a new mimesis scheme for partial observations, consisting of two strategies; (1) motion understanding from partial observations and (2) proto-symbol-based motion duplication. With the proposed method, whole-body motion imitation is possible even when observing partial motion data. The scheme enables a humanoid robot to imitate a new observed motion by utilizing its own prior knowledge, without learning the demonstrated motion. Evaluation factors, such as inheritance coordinate and matching error, are introduced to evaluate imitation performance. The feasibility of the proposed scheme is demonstrated by an evaluation for a 20-degree-of-freedom humanoid robot.

Stability is a challenging issue when controlling haptic interaction systems because unstable behavior may injure human operators or deteriorate the realism of the provided cues. Based on the passivity condition for sampled-data haptic systems, we propose a novel energy-bounding algorithm to ensure robustly stable haptic interactions. The proposed algorithm limits the energy generated by a sample-and-hold operator within the energy consumable by the effective damping elements in the haptic system. The algorithm also ensures that the VE and controller are passive. This guarantees robustly stable haptic interactions, regardless of the sampling frequency and its variations, but compromises the displayable impedance range of the VE. Simulations and experiments using a commercial haptic device are used to demonstrate the effectiveness of the proposed algorithm.

The challenge of persistent navigation and mapping is to develop an autonomous robot system that can simultaneously localize, map and navigate over the lifetime of the robot with little or no human intervention. Most solutions to the simultaneous localization and mapping (SLAM) problem aim to produce highly accurate maps of areas that are assumed to be static. In contrast, solutions for persistent navigation and mapping must produce reliable goal-directed navigation outcomes in an environment that is assumed to be in constant flux. We investigate the persistent navigation and mapping problem in the context of an autonomous robot that performs mock deliveries in a working office environment over a two-week period. The solution was based on the biologically inspired visual SLAM system, RatSLAM. RatSLAM performed SLAM continuously while interacting with global and local navigation systems, and a task selection module that selected between exploration, delivery, and recharging modes. The robot performed 1,143 delivery tasks to 11 different locations with only one delivery failure (from which it recovered), traveled a total distance of more than 40 km over 37 hours of active operation, and recharged autonomously a total of 23 times.

In this paper, we present a probabilistic analysis approach for analyzing market-based algorithms applied to the initial formation problem. These algorithms determine an assignment scheme for associating individual robots with goal positions necessary to achieve a desired formation while minimizing an objective function. The main contribution of this paper is a method that calculates the expected value of the objective function, which allows us to estimate and compare theoretically the performance of two task allocation algorithms. This probabilistic analysis is applied in different runtime scenarios. We validate our approach through both simulations and experiments with real robots.

The bounding gait for the Platform for Ambulating Wheels (PAW), a new and unique hybrid wheeled–leg system is presented. Two hypotheses are tested and discussed: first, that the robot's forward speed can be increased by increasing the leg liftoff angles and, second, that the addition of distally mounted actuated wheels can be used in running gaits such as the bound. Both hypotheses were tested experimentally and found to be valid.

We present a method for catheter localization in the left atrium based on the unscented particle filter (UPF), a Monte Carlo method employed in stochastic state estimation. Using an intracardiac echo (ICE) ultrasound catheter, we acquire ultrasound images of the atrium from multiple configurations and iteratively determine the catheter's pose with respect to anatomy. At each time step, the catheter's change in pose is determined using either a six-degree-of-freedom electromagnetic pose sensor or a robotic guide catheter whose kinematics serve as a pseudo-pose measurement. Sensor and kinematic model uncertainties are explicitly considered when computing the localization estimate. Acquired ultrasound images are compared with simulated ultrasound images based on segmented computed tomography (CT) or magnetic resonance (MR) data of the left atrium. The results of these comparisons are used to refine the localization estimate. After considering less than 30 seconds' worth of ICE data, our algorithm converges to an accurate pose estimate. Furthermore, our algorithm is robust to sensor drift and kinematic model errors, as well as gradual, unmodeled movements in the anatomy. Such problems typically complicate traditional image-based localization.

In this paper we present a framework that enables an operator to haptically and visually interact with a simulated dynamic environment subject to virtual holonomic constraints. The framework combines a geometric constraint solver with a constrained dynamics simulation engine that controls an admittance-type haptic display. This system takes on relevant issues in the context of assisted teleoperated tasks, from providing an intuitive interface for creating and combining virtual constraints, to haptically displaying rigid motion constraints in simulated environments subject to desired inertial dynamics. Two experiments carried out using the Cobotic Hand Controller haptic display are presented.

In this paper we develop the concept of reduction-based control, which is founded on a controlled form of geometric reduction known as functional Routhian reduction. We prove a geometric property of general serial-chain robots termed recursive cyclicity, identifying the inherent robot symmetries that we exploit with the Subrobot Theorem. This shows that any serial-chain robot can be decomposed for arbitrarily lower-dimensional analysis and control. We apply this method to construct stable directional three-dimensional walking gaits for a four-degree-of-freedom hipped bipedal robot. The controlled reduction decouples the biped's sagittal-plane motion from the yaw and lean modes, and on the sagittal subsystem we use passivity-based control to produce known planar limit cycles on flat ground. The unstable yaw and lean modes are separately controlled to 2-periodic orbits through their shaped momenta. We numerically verify the existence of stable 2-periodic straight-walking limit cycles and demonstrate turning capabilities for the controlled biped.

Our approach factors the mapping problem into natural sub-goals: building a metrical representation for local small-scale spaces; finding a topological map that represents the qualitative structure of large-scale space; and (when necessary) constructing a metrical representation for large-scale space using the skeleton provided by the topological map. We describe how to abstract a symbolic description of the robot's immediate surround from local metrical models, how to combine these local symbolic models in order to build global symbolic models, and how to create a globally consistent metrical map from a topological skeleton by connecting local frames of reference.

Several recent studies have investigated the problem of object feature extraction with artificial whiskers. Many of these studies have used an approach in which the whisker is rotated against the object through a small angle. Each small-angle "tap" of the whisker provides information about the radial distance between the base of the whisker and the object. By tapping at various points on the object, a full representation of the surface can be gradually constructed in three-dimensional space. It is clear, however, that this tapping method does not exploit useful information about object contours that could be extracted by "sweeping" the whisker against the object. Rotating the whisker against the object through a large angle permits the collection of a sequence of contact points as the whisker slips along the surface. The present paper derives an algorithm based on a numerical cantilever beam model of the whisker to measure object profile shape over a single large-angle whisker rotation using only information about torque and angle at the whisker base. The algorithm is validated experimentally using three different object shapes. As the method does not require measurement of force, it is well suited for implementation on an array of robotic whiskers to accomplish quick and precise object feature extraction.