IROS 2019 论文列表

We present a method to predict humans in unexplored map areas given limited observations of the environment. We used a geometric representation of the environment based on cost maps and semantic room categorization. Human density distributions were generated using a human tracker based on LiDAR data recorded by a mobile robot. A Gaussian Process (GP) regression model was created to predict human density in surrounding unobserved map locations. GP prediction performance was evaluated on density data recorded in a series of ten simulations of 25 persons in an office setting, and in real-world robot deployments in an office-like environment. Experimental results demonstrate that the current method can predict human locations with an accuracy average of 70%.

In this paper, we address the visibility-based target tracking problem in which a mobile observer moving along a p-route, which we define as a fixed path for target tracking, tries to keep a mobile target in its field-of-view. By drawing a connection to the watchman's route problem, we find a set of conditions that must be satisfied by the p-route. Then we propose a metric for tracking to estimate a sufficient speed for the observer given the geometry of the environment. We show that the problem of finding the p-route on which the observer requires minimum speed is computationally intractable. We present a technique to find a p-route on which the observer needs at most twice the minimum speed to track the intruder and a reactive motion strategy for the observer.

Solving a thorny issue of real-time path planning for surgery robot in uncertain environments, a novel algorithm named bidirectional continuous tree search (BCTS) is proposed. Most partially observable markov decision process (POMDP) planners address challenges of unknown environments with discrete states, observations and actions, which are fail to automate the operative procedure. However, the BCTS method addresses the issue by handling POMDPs in continuous state, observation and action spaces. The proposed approach has a bidirectional search structure with the intent of greatly improving the calculation efficiency. Meanwhile, Bayesian optimization (BO) algorithm is considered to dynamically sample promising actions while we construct a belief tree. In view of the speed of BO process, the upper and lower bounds of the optimal action values given by fast informed bound (FIB) and point-based value iteration (PBVI) limit the search scope, so we can improve the speed of BO. In addition, we apply an optimal path planning generator, radial basis function neural network (RBFNN), to obtain a smoother trajectory. Finally, simulation of glaucoma surgery has been carried out to explore the best surgical approach. The results show that the introduced structure can effectively guide the surgery robot to perform surgical procedures and receive a real-time as well as smooth path.

We propose a novel fast simulation framework for soft objects/robots with intermittent contacts, whose contact areas/locations can be varying. We first perform a balanced model reduction of the full-order FEM (finite element method) model for each contact mode with the contact forcing as the input and the shape of the object/robot as the output. We then devise the strategy of passive model reduction and passive model switching of these reduced-order models (each with its contact mode) utilizing the techniques of our recently-proposed passive mid-point integration (for the passivity of each reduced-order model) and simultaneous diagonalization (for the passivity of model reduction and model switching). The efficacy of the theory is then demonstrated with simulation and experimental results.

The underwater environment poses significant challenges for accurate autonomous underwater vehicle (AUV) navigation. Electromagnetic (EM) waves rapidly attenuate due to absorption by water, thereby preventing the use of traditional EM -based positioning methods such as Global Positioning System (GPS) or visible -light cameras. Consequently, underwater positioning is often performed using systems that operate in the hydro -acoustic frequency range (<; 10 MHz). Recent work has demonstrated the efficacy of a novel acoustic positioning approach for multi-AUV operations called passive inverted ultra -short baseline (piUSBL) localization - with each vehicle equipped with a time -synchronized USBL array, oneway travel -time (OWTT) range and angle between the AUV and a single acoustic beacon enables multi-AUV navigation relative to the beacon. In this work, a piUSBL system using a fi ve-hydrophone pyramidal array implemented on a WAM-V autonomous surface vehicle (ASV) was used to experimentally gather acoustic measurements and to compare the accuracy of piUSBL localization against ground -truth from a differential GPS unit. This paper provides a comprehensive analysis of the positioning accuracy of the system in a real -world environment, both prior to and after Bayesian fi ltering, using two independent acoustic beacons for validation. We demonstrate that piUSBL provides acoustic range and angle measurements with errors of about p, ± cr = 0.03 ± 1.49 m and p, ± cr = -0.11 ± 3.16 ° respectively. These experimental results suggest that piUSBL localization can provide a highly accurate, inexpensive, and low power navigation solution for the next generation of miniature, low-cost underwater vehicle.

Caging grasps limit the mobility of an object to a bounded component of configuration space. We introduce a notion of partial cage quality based on maximal clearance of an escaping path. As this is a computationally demanding task even in a two-dimensional scenario, we propose a deep learning approach. We design two convolutional neural networks and construct a pipeline for real-time partial cage quality estimation directly from 2D images of object models and planar caging tools. One neural network, CageMaskNN, is used to identify caging tool locations that can support partial cages, while a second network that we call CageClearanceNN is trained to predict the quality of those configurations. A dataset of 3811 images of objects and more than 19 million caging tool configurations is used to train and evaluate these networks on previously unseen objects and caging tool configurations. Furthermore, the networks are trained jointly on configurations for both 3 and 4 caging tool configurations whose shape varies along a 1-parameter family of increasing elongation. In experiments, we study how the networks' performance depends on the size of the training dataset, as well as how to efficiently deal with unevenly distributed training data. In further analysis, we show that the evaluation pipeline can approximately identify connected regions of successful caging tool placements and we evaluate the continuity of the cage quality score evaluation along caging tool trajectories. Experiments show that evaluation of a given configuration on a GeForce GTX 1080 GPU takes less than 6 ms.

In this paper, a novel method to construct a versatile gripper which can grasp various sizes and shapes of objects is proposed. The design and construction of this gripper is based on a paper folding technique called `modular origami' This particular type of origami is constructed from multiple modules which can transform its shape by rotation. The gripper is actuated by a single motor using a linkage mechanism. The grasping part of this gripper is made of paper, thus it provides sufficient compliance to bend itself and create multiple contact points around objects with different shapes. This gripper can be used in two modes: force grasping and cage grasping. The proposed gripper was tested on objects with various shapes, weights and sizes. Furthermore, how the property of the paper affects the performance of the gripper was also tested. The thickness of the gripper material affects the compliance of the gripper and the range of sizes of objects which can be successfully gripped.

We propose PanopticFusion, a novel online volumetric semantic mapping system at the level of stuff and things. In contrast to previous semantic mapping systems, PanopticFusion is able to densely predict class labels of a background region (stuff) and individually segment arbitrary foreground objects (things). In addition, our system has the capability to reconstruct a large-scale scene and extract a labeled mesh thanks to its use of a spatially hashed volumetric map representation. Our system first predicts pixel-wise panoptic labels (class labels for stuff regions and instance IDs for thing regions) for incoming RGB frames by fusing 2D semantic and instance segmentation outputs. The predicted panoptic labels are integrated into the volumetric map together with depth measurements while keeping the consistency of the instance IDs, which could vary frame to frame, by referring to the 3D map at that moment. In addition, we construct a fully connected conditional random field (CRF) model with respect to panoptic labels for map regularization. For online CRF inference, we propose a novel unary potential approximation and a map division strategy. We evaluated the performance of our system on the ScanNet (v2) dataset. PanopticFusion outperformed or compared with state-of-the-art offline 3D DNN methods in both semantic and instance segmentation benchmarks. Also, we demonstrate a promising augmented reality application using a 3D panoptic map generated by the proposed system.

Accurate object 6D pose estimation is a core task for robot bin-picking applications, especially when objects are randomly stacked with heavy occlusion. To address this problem, this paper proposes a simple but novel Point-wise Pose Regression Network (PPR-Net). For each point in the point cloud, the network regresses a 6D pose of the object instance that the point belongs to. We argue that the regressed poses of points from the same object instance should be located closely in pose space. Thus, these points can be clustered into different instances and their corresponding objects' 6D poses can be estimated simultaneously. In our experiments, PPR-Net outperforms the state-of-the-art approach by 15% - 41% in average precision when evaluated on the benchmark Siléane dataset. In addition, it also works well in real world robot bin-picking tasks.

In this paper, inspired by Proportional-Derivative (PD) control laws, we present a class of Control Lyapunov Function (CLF) based Quadratic Programs (QPs) for robotic systems. Proportional-Derivative (PD) control laws are independent of the robot model, however, they fail to incorporate physical constraints, such as torque saturation. On the other hand, most optimization based control design approaches ensure satisfaction of the physical constraints, but they are sensitive to errors in the robot model. The PD based Quadratic Programs (PD-QPs), presented in this paper, are a first step towards bridging this gap between the PD and the optimization based controllers to bring the best of both together. We derive two versions of PD-QPs: model-based and model-free. Furthermore, for tracking time-varying trajectories, we establish asymptotic stability for the model-based PD-QP, and ultimate boundedness for the model-free PD-QP. The performance of the PD-QPs is evaluated on two robot models: a fully actuated cart-pole and an underactuated 5-DOF biped.

In this paper, we propose PASS3D to achieve point-wise semantic segmentation for 3D point cloud. Our framework combines the efficiency of traditional geometric methods with robustness of deep learning methods, consisting of two stages: At stage -1, our accelerated cluster proposal algorithm will generate refined cluster proposals by segmenting point clouds without ground, capable of generating less redundant proposals with higher recall in an extremely short time; stage -2 we will amplify and further process these proposals by a neural network to estimate semantic label for each point and meanwhile propose a novel data augmentation method to enhance the network's recognition capability for all categories especially for non-rigid objects. Evaluated on KITTI raw dataset, PASS3D stands out against the state-of-the-art on some results, making itself competent to 3D perception in autonomous driving system. Our source code will be open-sourced. A video demonstration is available at https://www.youtube.com/watch?v=cukEqDuP_Qw.

Contemporary humanoids are equipped with visual and LiDAR sensors that are effectively utilized for Visual Odometry (VO) and LiDAR Odometry (LO). Unfortunately, such measurements commonly suffer from outliers in a dynamic environment, since frequently it is assumed that only the robot is in motion and the world is static. To this end, robust state estimation schemes are mandatory in order for humanoids to symbiotically co-exist with humans in their daily dynamic environments. In this article, the robust Gaussian Error-State Kalman Filter for humanoid robot locomotion is presented. The introduced method automatically detects and rejects outliers without relying on any prior knowledge on measurement distributions or finely tuned thresholds. Subsequently, the proposed method is quantitatively and qualitatively assessed in realistic conditions with the full-size humanoid robot WALK-MAN v2.0 and the mini-size humanoid robot NAO to demonstrate its accuracy and robustness when outlier VOLO measurements are present. Finally, in order to reinforce further research endeavours, our implementation is released as an open-source ROS/C++package.

Spatial perception is the backbone of many robotics applications, and spans a broad range of research problems, including localization and mapping, point cloud alignment, and relative pose estimation from camera images. Robust spatial perception is jeopardized by the presence of incorrect data association, and in general, outliers. Although techniques to handle outliers do exist, they can fail in unpredictable manners (e.g., RANSAC, robust estimators), or can have exponential runtime (e.g., branch-and-bound). In this paper, we advance the state of the art in outlier rejection by making three contributions. First, we show that even a simple linear instance of outlier rejection is inapproximable: in the worst-case one cannot design a quasi-polynomial time algorithm that computes an approximate solution efficiently. Our second contribution is to provide the first per-instance sub-optimality bounds to assess the approximation quality of a given outlier rejection outcome. Our third contribution is to propose a simple general-purpose algorithm, named adaptive trimming, to remove outliers. Our algorithm leverages recently-proposed global solvers that are able to solve outlier-free problems, and iteratively removes measurements with large errors. We demonstrate the proposed algorithm on three spatial perception problems: 3D registration, two-view geometry, and SLAM. The results show that our algorithm outperforms several state-of-the-art methods across applications while being a general-purpose method.

We tackle the INS/GPS sensor fusion problem for pose estimation, particularly in the common setting where the INS components (IMU and magnetometer) function at much higher frequencies than GPS, and where the magnetometer and GPS are prone to giving erroneous measurements (outliers) due to magnetic disturbances and glitches. Our main contribution is a novel non-linear optimization framework that (1) fuses pre-integrated IMU and magnetometer measurements with GPS, in a manner that respects the manifold structure of the state space; and (2) supports the usage of robust norms and efficient large scale optimization to effectively mitigate the effects of outliers. Through extensive experiments, we demonstrate the superior accuracy and robustness of our approach over filtering methods (which are customarily applied in the target setting) with minimal impact to computational efficiency. Our work further illustrates the strength of optimization approaches in state estimation problems and paves the way for their adoption in the control and navigation communities.

A Hybrid passive Linear Inverted Pendulum (H-LIP) model is proposed for characterizing, stabilizing and composing periodic orbits for 3D underactuated bipedal walking. Specifically, Period-l (P1) and Period -2 (P2) orbits are geometrically characterized in the state space of the H-LIP. Stepping controllers are designed for global stabilization of the orbits. Valid ranges of the gains and their optimality are derived. The optimal stepping controller is used to create and stabilize the walking of bipedal robots. An actuated Spring-loaded Inverted Pendulum (aSLIP) model and the underactuated robot Cassie are used for illustration. Both the aSLIP walking with PI or P2 orbits and the Cassie walking with all 3D compositions of the PI and P2 orbits can be smoothly generated and stabilized from a stepping-in-place motion. This approach provides a perspective and a methodology towards continuous gait generation and stabilization for 3D underactuated walking robots.

A Mobile Cable-Driven Parallel Robot (MCDPR) is composed of a classical Cable-Driven Parallel Robot (CDPR) carried by multiple mobile bases. The additional mobilities due the motion of the mobile bases allow such systems to autonomously modify their geometric architecture, and thus make them suitable for multiple manipulation tasks in constrained environments. Moreover, these additional mobilities mean MCDPRs are kinematically redundant and may use this redundancy to optimize secondary task criteria. However, the high dimensional state space and closed chain constraints add complexity to the motion planning problem. To overcome this, we propose a method for trajectory planning for MCDPRs performing pick and place operations in cluttered environments by using direct transcription optimization. Two different scenarios have been considered and their results are validated using a dynamic simulation software (V-REP) and experimentally.

Motion planning trajectories for a multi-limbed robot to climb up walls requires a unique combination of constraints on torque, contact force, and posture. This paper focuses on motion planning for one particular setup wherein a six-legged robot braces itself between two vertical walls and climbs vertically with end effectors that only use friction. Instead of motion planning with a single nonlinear programming (NLP) solver, we decoupled the problem into two parts with distinct physical meaning: torso postures and contact forces. The first part can be formulated as either a mixed-integer convex programming (MICP) or NLP problem, while the second part is formulated as a series of standard convex optimization problems. Variants of the two wall climbing problem e.g., obstacle avoidance, uneven surfaces, and angled walls, help verify the proposed method in simulation and experimentation.

In this paper, we address the problem of temporal logic planning given both hard specifications of the robot's mission and soft preferences on the plans that achieve the mission. In particular, we consider a problem whose inputs are a transition system, a linear temporal logic (LTL) formula specifying the robot's mission, and an ordered sequence of formulas expressed in linear dynamic logic over finite traces (LDLf) specifying the user's preferences for how the mission should be completed. The planner's objective is to synthesize, on this transition system, an infinite trajectory that best fits the user's preferences over finite prefixes of that trajectory while nonetheless satisfying the overall objective. We describe an algorithm for this problem that constructs, from the inputs, a product automaton -which is, in fact, a special kind of state-weighted Büchi automaton- over which an optimal trajectory is synthesized. This synthesis problem is solved via reduction to the minimax path problem in vertex weighted graphs, which can be solved by variants of the standard algorithms for computing shortest paths in a graph or by algorithms for the all-pairs bottleneck paths problem on vertex-weighted graphs. We show the applicability of the approach via some case studies, for which we present results computed by an implementation.

This paper addresses a multi-robot motion planning problem in probabilistic maps obtained by semantic simultaneous localization and mapping (SLAM). The goal of the robots is to accomplish complex collaborative high level tasks captured by global temporal logic specifications in the presence of uncertainty in the workspace. Specifically, the robots operate in an unknown environment modeled as a semantic map determined by Gaussian distributions over landmark positions and arbitrary discrete distributions over landmark classes. We extend Linear Temporal Logic by including information-based predicates allowing us to incorporate uncertainty and probabilistic satisfaction requirements directly into the task specification. We propose a new highly scalable sampling-based approach that synthesizes paths that satisfy the assigned task specification while minimizing a user-specified motion cost function. Finally, we show that the proposed algorithm is probabilistically complete, asymptotically optimal and supported by convergence rate bounds. We provide extensive simulation results that corroborate the theoretical analysis and show that the proposed algorithm can address large-scale planning tasks.