IROS 2017 论文列表

Developing surface classification models manually requires significant time and detracts from the goal of automating systems. We create a system that automatically collects the data using an Unmanned Aerial Vehicle (UAV), extracts features, trains a large number of classifiers, selects the best classifier, and programs the UAV with that classifier. Motivating our work is a prior project [1] that manually developed a surface classifier using an accelerometer; to verify our system functionality, we replicate those results with our new automated system and improve on those results, providing a four-surface classifier with a 75% classification rate and a hard/soft classifier with a 100% classification rate. We further verify our system through a field experiment that collects and classifies new data, proving its end-to-end functionality. Overall, our system reduces the time and machine learning expertise needed by the user to develop new time-series classifiers usable by the UAV. The general form of our system provides a valuable tool for automation of classifier creation and is released as an open-source tool [2].

We present a system for creating object models from RGB-D views acquired autonomously by a mobile robot. We create high-quality textured meshes of the objects by approximating the underlying geometry with a Poisson surface. Our system employs two optimization steps, first registering the views spatially based on image features, and second aligning the RGB images to maximize photometric consistency with respect to the reconstructed mesh. We show that the resulting models can be used robustly for recognition by training a Convolutional Neural Network (CNN) on images rendered from the reconstructed meshes. We perform experiments on data collected autonomously by a mobile robot both in controlled and uncontrolled scenarios. We compare quantitatively and qualitatively to previous work to validate our approach.

Dynamic camera cluster (DCC) calibration determines a time-varying set of extrinsic calibration transformations between cameras in a multi-camera cluster, where one or more cameras are mounted on actuated mechanisms. In this paper, we present a novel active vision approach for DCC calibration, which directly reduces the parameter uncertainty by selecting calibration measurements using an information theoretic next-best-view policy. Our system automatically selects the next best measurement for the calibration by determining the optimal actuator inputs which minimize the predicted covariance of the extrinsic parameters. We show that our method is able to successfully estimate calibration parameters up to a user specified accuracy with no manual excitation. We test our method in simulation on a variety of actuated mechanisms and validate the results on a real 3 axis gimbal, and demonstrate our approach is able to achieve accurate calibrations using fewer measurement sets when compared to existing approaches.

We present AutonoVi, a novel algorithm for autonomous vehicle navigation that supports dynamic maneuvers and integrates traffic constraints and norms. Our approach is based on optimization-based maneuver planning that supports dynamic lane-changes, swerving, and braking in all traffic scenarios and guides the vehicle to its goal position. We take into account various traffic constraints, including collision avoidance with other vehicles, pedestrians, and cyclists using control velocity obstacles. We use a data-driven approach to model the vehicle dynamics for control and collision avoidance. Furthermore, our trajectory computation algorithm takes into account traffic rules and behaviors, such as stopping at intersections and stoplights, based on an arc-spline representation. We have evaluated our algorithm in a simulated environment and tested its interactive performance in urban and highway driving scenarios with tens of vehicles, pedestrians, and cyclists. These scenarios include jaywalking pedestrians, sudden stops from high speeds, safely passing cyclists, a vehicle suddenly swerving into the roadway, and high-density traffic where the vehicle must change lanes to progress more effectively.

Robot software controllers are often concurrent and time critical, and requires modern engineering approaches for validation and verification. With this motivation, we have developed a tool and techniques for graphical modelling with support for automatic generation of underlying mathematical definitions for model checking. It is possible to check automatically both general properties, like absence of deadlock, and specific application properties. We cater both for timed and untimed modelling and verification. Our approach has been tried in examples used in a variety of robotic applications.

Recently in tandem with the spread of portable devices for reading electronic books, devices for digitizing paper books, called book scanners, are developed to meet the increased demand for digitizing privately owned books. However, conventional book scanners still have complex components to mechanically turn pages and to rectify the acquired images that are inevitably distorted by the curvy book surface. Here, we present the multi-scale mechanism that turns pages electronically using electroadhesive force generated by a micro-scale structure. Its another advantage is that perspective correction of image processing is applicable to readily reconstruct the distorted images of pages. Specifically, to turn one page at a time not two pages, we employ a micro-scale structure to generate near-field electroadhesive force that decays rapidly and accordingly attracts objects within tens of micrometers. We analyze geometrical parameters of the micro-scale structure to improve the decay characteristics. We find that the decay characteristics of electroadhesive force definitely depends upon the geometrical period of the micro-scale structure, while its magnitude depends on a variety of parameters. Based on this observation, we propose a novel electrode configuration with improved decay characteristics. Dynamical stability and kinematic requirements are also examined to successfully introduce near-field electroadhesive force into our digitizing process.

Selective weeding is one of the key challenges in the field of agriculture robotics. To accomplish this task, a farm robot should be able to accurately detect plants and to distinguish them between crop and weeds. Most of the promising state-of-the-art approaches make use of appearance-based models trained on large annotated datasets. Unfortunately, creating large agricultural datasets with pixel-level annotations is an extremely time consuming task, actually penalizing the usage of data-driven techniques. In this paper, we face this problem by proposing a novel and effective approach that aims to dramatically minimize the human intervention needed to train the detection and classification algorithms. The idea is to procedurally generate large synthetic training datasets randomizing the key features of the target environment (i.e., crop and weed species, type of soil, light conditions). More specifically, by tuning these model parameters, and exploiting a few real-world textures, it is possible to render a large amount of realistic views of an artificial agricultural scenario with no effort. The generated data can be directly used to train the model or to supplement real-world images. We validate the proposed methodology by using as testbed a modern deep learning based image segmentation architecture. We compare the classification results obtained using both real and synthetic images as training data. The reported results confirm the effectiveness and the potentiality of our approach.

Autonomous mobile robots that use multiple depth sensors to perceive their environments, rely on extrinsic calibration to combine the individual views from each sensor into a single coherent view of the surroundings. Such extrinsic calibration is tedious to perform manually, and requires that specific scenes to calibrate. Current state of the art automatic approaches do not consider the content of scenes used for calibration, and thus are not robust to partially informative scenes in long-term deployments. In this paper, we present Delta-Calibration, an automated extrinsic calibration technique that takes into account the information in a scene for calibration. Delta-Calibration relies on constrained sensor motion to minimize the effects of desynchronization, and ego-motion estimation from each depth camera to detect significant changes in pose, which we term Delta-Transforms. We derive a solution to the extrinsic calibration using such Delta-Transforms taking into account uncertain axes of motion in the environment, and further infer necessary and sufficient conditions on the Delta-Transforms such that Delta-Calibration results in a unique, non-singular, and numerically stable extrinsic calibration. We present quantitative and qualitative results demonstrating the effectiveness of Delta-Calibration at computing extrinsic calibration over different arrangements of depth sensors.

To leverage soft hands to their full potential for grasping, we propose to design their morphology and control signals together. Considering both parameter domains makes it easier and faster to find solutions compared to fixing parameters of either domain. Additionally, the approach scales well to high-dimensional parameter spaces, which is a precondition to make automated co-design useful for soft hands. We further present an efficient simulator for simulating grasps with soft hands which is based on the SOFA framework and enables us to simulate more than a million grasps per day. These two complementary improvements promise a boost in the development of competent soft hands and their control in the future.

Batch-cell manipulation is a key technology in biological applications. Robotic manipulation has important significance to improve the operation success rate and reduce the technical threshold, but the problem of inefficiency still exists in batch-cell experiments. In this paper, an automated cell transportation system is designed for batch-cell manipulation. It has some technical aspects such as a cell groove to contain the cells, the micromanipulator and motor stage control methods, and computer vision algorithms. Since the cells are arranged in a line in the groove, the transportation system improves the efficiency of finding the cells in the petri dish. Furthermore, the minimum pressure to drag and release the cell are analyzed theoretically, so that the other cells will not be affected when manipulating one cell. The visual algorithms to detect the cell position and cell holding state are evaluated by porcine oocyte. Experimental results show both algorithm has high success rates: 96% and 100%. Finally, cell rotating experiments are introduced to verify the effectiveness of the transportation system. The average transfer efficiency has been improved by 20% compared to manual operation. The results show that this system can be used in many manipulations.

Ground-based teleoperation of robots in space is subject to time delays of several seconds or more. This leads to the use of model-based approaches, where the operator interacts with a model (simulation) of the remote environment and the remote robot attempts to reproduce the results of that interaction. But, it is also desirable for the operator to view (delayed) images from the remote scene. These images, however, are often from one or more monocular cameras mounted on the robot end-effector, which leads to several other problems: unintuitive teleoperation due to the eye-in-hand configuration, limited field of view, and lack of stereo visualization. We present an augmented virtuality interface for teleoperation, which can solve these problems by projecting the real camera images onto a registered 3D model of the environment and allowing the operator to select any desired viewpoint. This approach is suitable when there is at least a partial model of the environment, as is the case for satellite servicing. The proposed method begins with a video survey to register the 3D model to the physical environment, followed by a user interface that presents a stereo visualization of the model, augmented by projections of real camera images onto the model. We quantitatively and qualitatively compare the augmented virtuality images to real camera images taken from the same viewpoint and perform experiments to evaluate the efficacy of the augmented virtuality paradigm for teleoperation. The results suggest that this approach can improve operator situation awareness, potentially leading to better performance, especially when the camera views are unintuitive or limited.

The ability to direct visual attention is a fundamental skill for seeing robots. Attention comes in two flavours: the gaze direction (overt attention) and attention to a specific part of the current field of view (covert attention), of which the latter is the focus of the present study. Specifically, we study the effects of attentional masking within pre-trained deep neural networks for the purpose of handling ambiguous scenes containing multiple objects. We investigate several variants of attentional masking on partially pre-trained deep neural networks and evaluate the effects on classification performance and sensitivity to attention mask errors in multi-object scenes. We find that a combined scheme consisting of multi-level masking and blending provides the best trade-off between classification accuracy and insensitivity to masking errors. This proposed approach is denoted multilayer continuous-valued convolutional feature masking (MC-CFM). For reasonably accurate masks it can suppress the influence of distracting objects and reach comparable classification performance to unmasked recognition in cases without distractors.

In this work, we provide a solution for posturing the anthropomorphic Robonaut-2 hand and arm for grasping based on visual information. A mapping from visual features extracted from a convolutional neural network (CNN) to grasp points is learned. We demonstrate that a CNN pre-trained for image classification can be applied to a grasping task based on a small set of grasping examples. Our approach takes advantage of the hierarchical nature of the CNN by identifying features that capture the hierarchical support relations between filters in different CNN layers and locating their 3D positions by tracing activations backwards in the CNN. When this backward trace terminates in the RGB-D image, important manipulable structures are thereby localized. These features that reside in different layers of the CNN are then associated with controllers that engage different kinematic subchains in the hand/arm system for grasping. A grasping dataset is collected using demonstrated hand/object relationships for Robonaut-2 to evaluate the proposed approach in terms of the precision of the resulting preshape postures. We demonstrate that this approach outperforms baseline approaches in cluttered scenarios on the grasping dataset and a point cloud based approach on a grasping task using Robonaut-2.

Existing trajectory planning based locomotion algorithms lack the analytic tools to fully comprehend energy based movements that would allow for full stability and mobility. Such drawbacks make humanoid robots' locomotion sensitive to external disturbances and compromise robots' agility in unstructured environment. In this work, we specifically focus on the push recovery problem for humanoid robots. We propose an approach to design a nonlinear controller that is robust to external disturbances. It allows the state of the rigid body dynamics to asymptotically converge to the subspace that meet the criteria of balancing, based on the properties of artificial invariant subspace and potential functions. Our algorithm is completely adaptive in real time without requiring trajectory planning in advance. We demonstrate the robustness of the proposed algorithm base on extensive push recovery experiments on the DARWIN-OP robot platform on flat terrains.

We propose an algorithm called σ-approximation that compresses the non-zero values of beliefs for partially observable Markov decision processes (POMDPs) in order to improve performance and reduce memory usage. Specifically, we approximate individual belief vectors with a fixed bound on the number of non-zero values they may contain. We prove the correctness and a strong error bound when the σ-approximation is used with the point-based value iteration (PBVI) family algorithms. An analysis compares the algorithm on six larger domains, varying the number of non-zero values for the σ-approximation. Results clearly demonstrate that when the algorithm used with PBVI (σ-PBVI), we can achieve over an order of magnitude improvement. We ground our claims with a full robotic implementation for simultaneous navigation and localization using POMDPs with σ-PBVI.

Kinematic calibration of a robotic manipulator determines its fabricated dimensions that inevitably deviate, due to manufacturing tolerances, from the specified/designed geometry. Conventional calibration methods achieve this goal by deriving a system of constraint equations from the kinematic structure. Workspace coordinates and the corresponding joint space coordinates are experimentally obtained so that these constraint equations can be solved for the optimal geometric parameters that best fit the data. This paper proposes an alternative kinematic calibration method that leverages response surface methodology to obtain empirical models of direct and inverse kinematics from the same experimental data. The advantages the proposed method include nullifying the requirements of deriving constraint equations and performing joint sensor calibration. In addition, since the developed models are empirical in nature, whether closed-form solutions to the direct or the indirect model is available becomes conveniently inconsequential. Both physical and simulation experiments were conducted on a 3-PSS/S to evaluate the accuracy of the proposed method.

Accurate kinematic models and measurements are needed in many robotic applications. However uncertainties related to joint angle measurements and manipulator geometry are unavoidable, especially when grasping and using different tools or when we do not have access to an accurate robot model, e.g. when we construct a robotic system by hand. The generalized inverse kinematics methods are not applicable when a manipulator stay inside its singular region. We derive the upper bounds on the joint measurement errors and geometric uncertainties, in order to guarantee that the open-chain serial manipulators stay outside the singular region. These bounds in other words enable en effective execution of generalized inverse kinematics methods for a robotic system which is prone to geometric uncertainties. In addition to the analytic derivation, We validate the proposed bounds through a trajectory tracing task performed by a PR2 robot simulator.

This paper introduces an anytime synthesized motion planning algorithm for humanoid robots unifying locomotion and manipulation planning. It generates an entire set of motions to finish specific tasks in an environment containing obstacles by exploiting a powerful inverse kinematics (IK) engine. The IK engine can compute solutions allowing the robot to reposition its feet for meeting the task requirements. The presented planning algorithm has two primary beneficial capabilities. First, it is capable of generating a motion plan to complete a task handling multiple ordered or unordered actions. Second, it produces an initial solution very quickly, and then searches for the opportunity to improve the the solution during execution. The performance of the proposed algorithm is evaluated on the NASA-JSC Valkyrie humanoid robot by demonstrating an object pick up task in simulation and a box pick-and-place task in the real world.

Hybrid driving-stepping locomotion is an effective approach for navigating in a variety of environments. Long, sufficiently even distances can be quickly covered by driving while obstacles can be overcome by stepping. Our quadruped robot Momaro, with steerable pairs of wheels located at the end of each of its compliant legs, allows such locomotion. Planning respective paths attracted only little attention so far. We propose a navigation planning method which generates hybrid locomotion paths. The planner chooses driving mode whenever possible and takes into account the detailed robot footprint. If steps are required, the planner includes those. To accelerate planning, steps are planned first as abstract manoeuvres and are expanded afterwards into detailed motion sequences. Our method ensures at all times that the robot stays stable. Experiments show that the proposed planner is capable of providing paths in feasible time, even for challenging terrain.