ICRA 2019 论文列表

Registration accuracy is influenced by the presence of outliers and numerous robust solutions have been developed over the years to mitigate their effect. However, without a large scale comparison of solutions to filter outliers, it is becoming tedious to select an appropriate algorithm for a given application. This paper presents a comprehensive analysis of the effects of outlier filters on the Iterative Closest Point (ICP) algorithm aimed at a mobile robotic application. Fourteen of the most common outlier filters (such as M-estimators) have been tested in different types of environments, for a total of more than two million registrations. Furthermore, the influence of tuning parameters has been thoroughly explored. The experimental results show that most outlier filters have a similar performance if they are correctly tuned. Nonetheless, filters such as Var. Trim., Cauchy, and Cauchy MAD are more stable against different environment types. Interestingly, the simple norm L1 produces comparable accuracy, while being parameterless.

This paper investigates strategies for 3D multi-agent position control using a shared control input and self-propelled agents. The only control inputs allowed are rotation commands that rotate all agents by the same rotation matrix. In the 2D case, only two degrees-of-freedom (DOF) in position are controllable. We review controllability results in 2D, and then show that interesting things happen in 3D. We provide control laws for steering up to nine DOF in position, which can be mapped in various ways, including to control the x, y, z position of three agents, make four agents meet, or reduce the spread of n agents.

In this paper, a two-layer architecture for the bilateral teleoperation of multi-arms systems with communication delay is presented. We extend the single-master-single-slave two layer approach proposed in [1] by connecting multiple robots to a single energy tank. This allows to minimize the conservativeness due to passivity preservation and to increment the level of transparency that can be achieved. The proposed approach is implemented on a realistic surgical scenario developed within the EU-funded SARAS project.

Scene generation is an important step of robotic drawing. Recent works have shown success in scene generation conditioned on text using a variety of approaches, with which the generated scenes cannot be revised after its generation. To allow modification on generated scenes, we model the scene generation process as a discrete event system. Instead of training text-to-pixel mappings using large datasets, the proposed approach uses object instances retrieved from the Internet to synthesize scenes. Evaluated on 128 experiments using MSCOCO evaluation dataset, the result shows the scene generation performance has been increased by 197%, 22.3%, and 55.7% compared with the state of the art approach on three standard metrics (CIDEr, ROUGH-L, METEOR), respectively. Human evaluation conducted on Amazon Mechanical Turk shows over 80% of generated scenes are considered to have higher recognizability and better alignment with natural language descriptions than baseline works.

Navigation and control are a largely unsolved problems for micro autonomous underwater vehicles (μAUVs). The main challenges are due to the lack of accurate underwater localization systems, which fit on-board of μAUVs. In this work, we present an integrated navigation and control architecture consisting of a low-cost embedded localization module and an underwater way-point tracking controller, which fulfills the requirements of μAUVs. The performance of the navigation and control system is benchmarked in two different experimental scenarios.

This paper proposes a new control-oriented model to compute the magnetic field created by a coil. A major challenge for untethered microscale mobile robotics is the control of objects for precise and fast displacements. In this work, we propose to use an alternative implementation of a model based on elliptic integral functions to control magnetically actuated micro-robots. It allows to compute the magnetic field even in the area close to the coil quickly and accurately. This model is evaluated numerically and compared to classical approaches - dipole approximation, map-based interpolation and classical elliptic integral models - in terms of accuracy, computation time and memory requirement. Simulation results show that this works allows to have an accurate model in the whole workspace by avoiding numerical issues encountered in previous works. It can be computed in a few milliseconds, making it the right candidate for closed-loop control of magnetically actuated micro-robots.

We present a novel open-source tool for extrinsic calibration of radar, camera and lidar. Unlike currently available offerings, our tool facilitates joint extrinsic calibration of all three sensing modalities on multiple measurements. Furthermore, our calibration target design extends existing work to obtain simultaneous measurements for all these modalities. We study how various factors of the calibration procedure affect the outcome on real multi-modal measurements of the target. Three different configurations of the optimization criterion are considered, namely using error terms for a minimal amount of sensor pairs, or using terms for all sensor pairs with additional loop closure constraints, or by adding terms for structure estimation in a probabilistic model. The experiments further evaluate how the number of calibration boards affect calibration performance, and robustness against different levels of zero mean Gaussian noise. Our results show that all configurations achieve good results for lidar to camera errors and that fully connected pose estimation shows the best performance for lidar to radar errors when more than five board locations are used.

The recent proliferation of low-cost Micro Aerial Vehicles (MAV) offers an attractive means for inspecting critical infrastructure autonomously. However, to enable such autonomous tasks requires a precise spatial model of the structure and operational area, typically constructed using sensor measurements obtained from the environment. To facilitate autonomous inspection capabilities, we address the problem of autonomous MAV exploration and coverage of an unknown structure to acquire the spatial information necessary for the development of a high-fidelity 3D model of the structure. Key to this problem is to not only cover the entire structure to acquire a complete set of spatial measurements, but also to minimize accumulative data errors during the exploration through direct planning of loop closures. We introduce a real-time waypoint planning approach to guide MAV motions to achieve complete exploration, coverage, and loop closure while respecting limited onboard resources.

Lower-limb exoskeletons are of great interest in the robotics community because of their various applications in enhancement and rehabilitation. In this paper we present an autonomous exoskeleton platform for ankle plantarflexion assistance. The untethered exoskeleton has a high efficiency transmission system with reduction ratio of 27.4:1. This allows relocating the actuator to the wearer's hip, which reduces device inertia. A feed-forward controller based on field oriented control was implemented to control the brushless DC motor on the exoskeleton. Through various performance tests, the exoskeleton was shown to provide a torque control bandwidth of 17.5Hz and can effectively track biological torque profiles. The augmentation factor (AF) of the exoskeleton is 64.7W, implying potential to reduce walking metabolic cost. This exoskeleton establishes an autonomous platform for experiments involving ankle assistance.

This paper presents a pipeline for semantic segmentation of trees into their components. Given a single RGB-D image of a tree, we employ a deep network to predict labels to classify each pixel of the tree into trunk, branches, twigs and leaves. Multiple convolutional neural network architectures to combine the complementary modalities of depth and colour data are investigated. An asynchronous training approach where two networks trained separately on RGB and depth encoded as a 3-channel HHA image are combined using a late fusion architecture with different learning rates performs the best. Training and evaluation are performed on a synthetic dataset of 6 species of broadleaf trees. We further demonstrate the network's generalization capabilities, across various tree species on the synthetic dataset, achieving an accuracy of upto 92.5%. Furthermore, we present a qualitative evaluation of our approach on real-world data.

Finding sources of airborne chemicals with mobile sensing systems finds applications across the security, safety, domestic, medical, and environmental domains. In this paper, we present an algorithm based on source term estimation for odor source localization that is coupled with a navigation method based on partially observable Markov decision processes. We propose an innovative strategy to balance exploration and exploitation in navigation. The method has been evaluated systematically through high-fidelity simulations and in a wind tunnel emulating realistic and repeatable conditions. The impact of multiple algorithmic and environmental parameters has been studied in the experiments.

Current robotics industry trends show an increased interest in the interaction between humans and robots in a variety of fields, ranging from collaborative robots in manufacturing to assisted medical devices in the medical field. One limiting factor in present applications is the ability to actively morph these robotic structures and control their stiffness using the same type of actuation system. This paper focuses on developing an actively controlled, variable stiffness structure that uses a pneumatic system for both morphing and locking the structure shape. The structure design integrates Pneumatic Artificial Muscles (PAMs) that are pressurized to control shape morphing. The pressurization of the PAM provides a radial force that allows bi-directional morphing based on the pressurization scheme. Layer Jamming, which utilizes varied friction between thin sheets based on pressure, is used to control the variable stiffness of the structure. In this paper, a control model is developed to predict the morphed curvature of the structure based on the input actuator pressure. This experimental control model is also validated using a theoretical pseudo-rigid-body model. The repeatability and accuracy of morphing is also discussed. Through experimental testing, a measure of the stiffness variation range of the structure is also developed. This novel research would positively impact the robotics field by creating lightweight morphing structures that are flexible and easily deformed, but also stiff with high load-carrying capability for increased human-robot interaction.

In this paper, we present a method for depth control of one degree of freedom (1DOF) underwater robotic platform aMussel, based on the measurements from the ambient light sensor. Since ambient light values change during the day and depend on the weather conditions, references for the controller are acquired from other aMussel holding depth using pressure sensor based controller. Control inputs are transmitted using acoustic communication.

We present a differential-algebraic formulation with switching constraints to model the nonsmooth dynamics of robotic systems subject to changing constraints and multiple impacts. The formulation combines a single structurally simple governing equation, a set of switching kinematic constraints, and the plastic impact law, to represent the dynamics of robots that interact with their environment. The main contribution of this formulation is a novel algorithmic impact resolution method which provides an explicit solution to the classical plastic impact law in the case of multiple simultaneous impacts. This method serves as an alternative to prior linear-complementarity-based formulations which offer an implicit impact resolution through iterative calculation. We demonstrate the utility of the proposed method by simulating the locomotion of a planar anthropometric biped.

This paper introduces a probabilistic prediction model with a novel variant of the Markov decision process to improve tracking time and location detection accuracy in an air-to-ground robot surveillance scenario. While most surveillance algorithms focus mainly on controls of an unmanned aerial vehicle (UAV) and camera for faster tracking of an unmanned ground vehicle (UGV), this paper proposes a way of minimizing detection and tracking time by applying a prediction model to the first observed path taken by the UGV. We present a pursuit algorithm that addresses the problem of target (UGV) localization by combining prediction of used planning algorithm by the target, and application of the same planning algorithm to predict future trajectories. Our results show a high predictive accuracy based on a final position attained by the target and the location predicted by our model.

Extending our recent work [1] that focuses on the observability analysis of aided inertial navigation systems (INS) using homogeneous geometric features including points, lines and planes, in this paper, we complete the analysis for the general aided INS using different combinations of geometric features (i.e., points, lines and planes). We analytically show that the linearized aided INS with different feature combinations generally possesses the same observability properties as those with same features, i.e., 4 unobservable directions, corresponding to the global yaw rotation and the global position of the sensor platform. During the analysis, we particularly propose a novel minimal representation of line features, i.e., the “closest point” parameterization, which uses a 4D Euclidean vector to describe a line and is proved to preserve the same observability properties. Based on that, for the first time, we provide two sets of unified representations for points, lines and planes, i.e., the quaternion form and the closest point (CP) form, and perform extensive observability analysis with analytically-computed Jacobians for these unified parameterizations. We validate the proposed CP representations and observability analysis with Monte-Carlo simulations, in which EKF-based vision-aided INS (VINS) with combinations of geometrical features in CP form are developed and compared.

We propose a novel adsorption pad for wall climbing robot and irregular surface object using capillary force and water sealing. We call this adsorption pad as Super Wet Adsorption pad. The SWA pad has a porous part and a capillary part. The porous part is made by salt reaching method. When the SWA pad adsorbs to the wall which some sand and dust are attached, water comes from the porous part to avoid vacuum breaking. The capillary part is connected to the porous part to supply and stock the water. In this paper, we show the design of the porous part and the capillary part, fabrication process of each parts, and perform the evaluation experiment of the capillary force and adsorption of uneven surfaces, demonstration of wall climbing robot and adsorption of irregular surface foods.

We present an approach to safe physical Human-Robot Interaction (pHRI) for industrial robots, including collision detection, distinguishing accidental from intentional contacts, and achieving collaborative tasks. Typical industrial robots have a closed control architecture that accepts only velocity/position reference inputs, there are no joint torque sensors, and little or no information is available to the user on robot dynamics and on low-level joint controllers. Nonetheless, taking also advantage of the presence of a Force/Torque (F/T) sensor at the end-effector, a safe pHRI strategy based on kinematic information, on measurements from joint encoders and motor currents, and on end-effector forces/torques can be realized. An admittance control law has been implemented for collaboration in manual guidance mode, with whole-body collision detection in place both when the robot is in autonomous operation and when is simultaneously collaborating with a human. Several pHRI experiments validate the approach on a KUKA KR5 Sixx R650 robot equipped with an ATI F/T sensor.

Recent work in the design of mechanical systems for terrestrial locomotion has indicated successful strategies for increasing the energetic performance of a robotic locomotor without upgrading its actuator system. We apply one such strategy, termed power modulation, in a new way: for the design of a leg mechanism useful for running. Power modulation geometrically defines force/torque ratios between robot components to mechanically achieve certain energy transmission characteristics during fast stance dynamics that increase the kinetic power output of the overall system. Furthermore, we investigate the design of a leg mechanism that can adjust to exhibit power modulation. In this way, a leg mechanism would exhibit a low power mode for flat terrain, and can adjust to a high power mode for rough terrain. The latter makes jumping possible and extends the range of available footholds that can be accessed in a single step. To find a suitable leg mechanism, we leverage the Finite Root Generation method to compute a design. The design is advanced to a prototype and basic experiments are conducted to investigate its behavior as adjusted between high-and low-power modes.