IROS 2015 Papers

We present a novel cell manipulation method based on vibration-induced flow in open-chip environment. By applying circular vibration to a chip having micropillar patterns on its surface, local whirling flow is induced around the micropillars. Therefore, by patterning micropillar array on a chip and applying the circular vibration to the chip, we can achieve cell transportation along the array in an open chip environment. We theoretically analyzed this phenomenon with a single micropillar, and experimentally evaluated behaviours of manipulated microbeads with two micropillars. We confirm there are different manipulation modes according to the pitch of micropillars. We designed the micropillar array for an oocyte transportation based on these evaluations. We succeeded in the transportation of oocytes with up to 35.6 μm/s and confirmed that the velocity can be changed by changing the frequency and amplitude of the applied vibration.

Dynamic modeling of human bipedal walking is important for studying human locomotion and designing assistive and rehabilitation robotic devices. Physical principle-based models and data-driven learning models are two main methods to obtain human walking dynamics. We present analysis and connections between these two different modeling approaches. Mapping and correspondence are established between the reduced analytical dynamics (i.e., zero dynamics) of the bipedal walking and the learned latent dynamics from walking kinematic data. We present these mapping functions, their properties and applications. Experiments are presented to demonstrate and illustrate the analysis and results.

Natural language interfaces are powerful tools that enables humans and robots to convey information without the need for extensive training or complex graphical interfaces. Statistical techniques that employ probabilistic graphical models have proven effective at interpreting symbols that represent commands and observations for robot direction-following and object manipulation. A limitation of these approaches is their inefficiency in dealing with larger and more complex symbolic representations. Herein, we present a model for language understanding that uses parse trees and environment models both to learn the structure of probabilistic graphical models and to perform inference over this learned structure for symbol grounding. This model, called the Hierarchical Distributed Correspondence Graph (HDCG), exploits information about symbols that are expressed in the corpus to construct minimalist graphical models that are more efficient to search. In a series of comparative experiments, we demonstrate a significant improvement in efficiency without loss in accuracy over contemporary approaches for human-robot interaction.

After the incredible success of deep learning in the computer vision domain, there has been much interest in applying Convolutional Network (ConvNet) features in robotic fields such as visual navigation and SLAM. Unfortunately, there are fundamental differences and challenges involved. Computer vision datasets are very different in character to robotic camera data, real-time performance is essential, and performance priorities can be different. This paper comprehensively evaluates and compares the utility of three state-of-the-art ConvNets on the problems of particular relevance to navigation for robots; viewpoint-invariance and condition-invariance, and for the first time enables real-time place recognition performance using ConvNets with large maps by integrating a variety of existing (locality-sensitive hashing) and novel (semantic search space partitioning) optimization techniques. We present extensive experiments on four real world datasets cultivated to evaluate each of the specific challenges in place recognition. The results demonstrate that speed-ups of two orders of magnitude can be achieved with minimal accuracy degradation, enabling real-time performance. We confirm that networks trained for semantic place categorization also perform better at (specific) place recognition when faced with severe appearance changes and provide a reference for which networks and layers are optimal for different aspects of the place recognition problem.

In this paper we study performance achievable in delayed bilateral teleoperation control. Our goal is to characterize all teleoperation system behaviors that can be achieved with delayed communication but without any restrictive assumptions on the control law, such as passivity. To this end, a recent non-restrictive stabilizing control architecture for bilateral teleoperation is exploited. The main properties of this architecture are discussed and used to obtain parameterization of all the admittance matrices feasible under the nominal stability constraint. The use of this parameterization for shaping system behavior is demonstrated and the results are compared with those obtained using the classical passivity-based control schemes.

To design good cooperative team members in robotics it is important to know what coordination mechanisms are required. Our approach to explore the need for a coordination mechanism is based on a systematic methodology to identify team coordination requirements. We show that a team combined of robots that each individually can solve a task not always is able to guarantee task completion as a team. In these cases some mechanism for coordination is required and we formally identify various problem classes that impose different requirements. We introduce a formal task model and distinguish between no, implicit and explicit coordination mechanisms. This model is used to study which mechanisms guarantee task completion. It allows us to prove some empirical findings reported in the literature such as that a simple foraging task does not require coordination.

We consider in this paper cable driven parallel robot (CDPR) with m catenary cables that present a significant deformation due to their elasticity and own mass and we address the inverse kinematics (IK) of such a robot i.e. determining the lengths at rest of the cables for a given pose of the platform. We show that the IK requires to solve a square system of equations if the robot has up to 6 cables and an underconstrained system if the number of cables is larger than 6. In this paper we address the problem of finding all solutions of the IK when the number of cables is lower or equal to 6. We propose an algorithm that allows one to determine all possible solutions. for any number of cables. We illustrate the use of the IK solver on a CDPR with 6 cables and shows that the IK may have no solution even if the pose lie within the envelope of the base points or may have several solutions, in which case some cables may have a length at rest that is much larger than the distance between the anchor points of the cable.

Most cable-driven parallel robots are kinematically over-constrained mechanisms. This results in a non-trivial computation of the forward kinematic transformation. It is well known that the forward kinematics of parallel robots may have multiple solutions and in general the convergence of numerical methods is unknown. In recent works, it was proposed to formulate the forward kinematics as optimization problem that models the cables as linear springs in order to compute the platform pose which has minimal potential energy in the cables. In this paper, we analyzed this objective function. Using the Hessian matrix, we show that under certain conditions the problem at hand is convex and we can expect a unique and stable minimum. The computations are exemplified for point-shaped platforms as well as for the planar case. For the spatial case, we present an encouraging numerical study. An ordinary least squares method is then applied to find a position approximation and an improvement to previous methods is demonstrated.

In this paper a novel multi-modal tactile sensor is presented, featuring a matrix of capacitive pressure sensors, a microphone for acoustic measurements and proximity and ambient light sensor. The sensor is fully embedded and can be easily integrated at mechanical and electrical levels with industrial grippers. Tactile sensing design has been put on the same level of additional requirements, usually overlooked in tactile sensor research, such as the mechanical interface, cable harness and robustness against continuous and repetitive operations, just to name but a few. The performances of the different sensing modalities have been assessed in a test rig for tactile sensors. Experiments have been performed in order to show the capabilities of the sensor for implementing tactile based industrial gripper control and tactile based fabric classification.

This paper examines the problem of transitioning between gaits using a sagittal-plane reduced-order model of quadrupedal running, which has compliant elements in its torso and legs. First, periodic motions that correspond to pronking and two variations of bounding are generated. Analysis of the torso pitch motion in these gaits reveals basic differences that determine whether transitions between these gaits can be achieved. This observation is then incorporated in control design to enhance the stability of the generated motions. The domain of attraction of the closed-loop systems is estimated through simulation, showing that transition between pronking and bounding can be realized as a sequence of switchings between fixed points. The results in this work can be regarded as the first step towards synthesizing controllers for gait transitions within template models for quadrupedal running gaits.

This paper investigates the ability of dynamically walking bipeds to adapt their motion to persistent exogenous forcing. Applications that involve physical interactions between a bipedal robot and other robots (or humans), require that the robot adjust its stepping pattern in response to externally applied force signals. In our setting, an underactuated bipedal robot model walks under the influence of an external force. First, the hybrid zero dynamics method is used to design controllers that stabilize periodic walking motions in the absence of the external force. Then, conditions are derived analytically under which these (unforced) periodic gaits are modified to new (possibly aperiodic) stepping patterns that are consistent with the external force. It is deduced that underactuation holds the key to the ability of our model to adapt to external forces.

In this paper, we address the problem of optimal path planning to visit a set of regions by Dubins vehicle, which is also known as the Dubins Traveling Salesman Problem with Neighborhoods (DTSPN). This problem can be tackled by a transformation to other variants of the TSP or evolutionary algorithms. We address the DTSPN as a problem to find Dubins path to visit a given sequence of regions and propose a simple iterative optimization procedure to find Dubins path visiting the regions. The proposed approach allows to efficiently solve the DTSPN and based on the presented comparison with existing approaches, the proposed algorithm provides solutions of competitive quality to the evolutionary techniques while it is significantly less computationally demanding.

This paper presents a new control scheme for dexterous manipulation of an object by a multifingered hand. The aim of this control is to ensure the tracking by the object of specific trajectories depending on the task. More specifically, we consider a decoupling control according to the motion directions in the task space. The control scheme is based on a new eigenstructure assignment method, where the placement of the closed-loop eigenvalues ensures the stability of the system hand/object, and the strict placement of the object-level eigenvectors ensures the system decoupling. A parallel is made between task specifications and synergies. Current synergy controllers are often bio-inspired and based on the analysis of synergies in the human hand. The proposed algorithm allows the specification of precise object level performance according to the task, and to find the related synergies. Moreover, robustness to uncertainties on the contact point is an additional objective of the control law, and the effect of these uncertainties on the decoupling is investigated. Simulation results are presented and illustrate the decoupling effect of the control law on the system hand/object in the 2D-cartesian space.

The modeling and control of a vectoring tricopter UAV are developed in this article. The UAV is actuated by three thrust motors, each guided by suitable actuators, thus forming a platform able to independently track any desired attitude and trajectory. The derivation of the equations of motion is followed by the development of a vectoring controller that is supplemented by an allocation strategy. Both are based on geometric feedback linearization techniques, resulting in a singularity-free control law, taking into account the inertia effects of the main body, of the motors, and of the vectoring dynamics (actuators). A stability proof is developed validating the effectiveness of the control strategy under bounded disturbances. Simulations showcase the developed controller and tricopter performance.

Supervision and control of Autonomous underwater vehicles (AUVs) has traditionally been focused on an operator determining a priori the sequence of waypoints of a single vehicle for a mission. As AUVs become more ubiquitous as a scientific tool, we envision the need for controlling multiple vehicles which would impose less cognitive burden on the operator with a more abstract form of human-in-the-loop control. Such mixed-initiative methods in goal-oriented commanding are new for the oceanographic domain and we describe the motivations and preliminary experiments with multiple vehicles operating simultaneously in the water, using a shore-based automated planner.

In this work we introduce a low-level system that could be employed by a social robot like a robotic wheelchair or a humanoid, for approaching a group of interacting humans, in order to become a part of the interaction. Taking into account an interaction space that is created when at least two humans interact, a meeting point can be calculated where the robot should reach in order to equitably share space among the interacting group. We propose a sensor-based control task which uses the position and orientation of the humans with respect to the sensor as inputs, to reach the said meeting point while respecting spatial social constraints. Trials in simulation demonstrate the convergence of the control task and its capability as a low-level system for human-aware navigation.

For autonomous navigation of Micro Aerial Vehicles (MAVs) in cluttered environments, it is essential to detect potential obstacles not only in the direction of flight but in their entire local environment. While there exist systems that do vision based obstacle detection, most of them are limited to a single perception direction. Extending these systems to a multi-directional sensing approach would exhaust the payload limit in terms of weight and computational power. We present a novel light-weight sensor setup comprising of four stereo heads and an inertial measurement unit (IMU) to perform FPGA-based dense reconstruction for obstacle detection in all directions. As the data-rate scales up with the number of cameras we use an FPGA to perform streaming based tasks in real-time and show a light-weight polar-coordinate map to allow a companion computer to fully process the data of all the cameras and perform obstacle detection in real-time. The system is able to process up to 80 frames per second (fps) freely distributed on the four stereo heads while maintaining a low power budget. The perception system including FPGA, image sensors and stereo mounts is 235 g in weight.

In this paper the developed method for off-line compensation of tool deflections when milling aluminum with an industrial robot is presented. The efficiency of this approach is verified with high precision measurements of deflections using a laser tracker. The compensation method includes both the static milling process model which can predict the mean value components of the tool forces and a new combined local/global approach for estimating the combined stiffnesses of joints. With a process model such as the one presented in this paper and estimates of the robot's joint stiffness values, the tool path can be adjusted to counteract deflections of the tool during milling operations. The model was estimated from a large set of machining experiments and is valid for a given tool and material. The stiffness of each joint is a combination of several effects: (i) stiffness of the links, (ii) stiffness of joint bearings and gears and (iii) stiffness of the position control loops given by the individual axis controller gains in the controller software. The method presented in this paper increases accuracy when using an industrial robot for milling in relatively hard materials.

Large debris created after natural disasters restrict access to inner parts of affected regions and slows down disaster relief operations. Power shovels are often used to clear wreckage but the process can take a lot of time. Moreover, it is dangerous to involve human workers in such conditions due to loose debris collapsing and causing severe damage. To speedup access to inner areas, instead of removing obstacles, they can be surmounted with the assistance of a carefully maneuvered power shovel arm. In this paper, an autonomous obstacle surmounting technique for an unmanned power shovel is proposed. Out of different sequences, the one that optimizes the total energy consumption is selected as the best candidate for surmounting a given obstacle by dynamic simulations. The experiments conducted using a miniaturized power shovel show the effectiveness of the proposed method.