ICRA 2018 论文列表

Recently, the avenue of adaptable, soft robotic hands has opened simplified opportunities to grasp different items; however, the potential of soft end effectors (SEEs) is still largely unexplored, especially in human-robot interaction. In this paper, we propose, for the first time, a simple touch-based approach to endow a SEE with autonomous grasp sensory-motor primitives, in response to an item passed to the robot by a human (human-to-robot handover). We capitalize on human inspiration and minimalistic sensing, while hand adaptability is exploited to generalize grasp response to different objects. We consider the Pisa/IIT SoftHand (SH), an under-actuated soft anthropomorphic robotic hand, which is mounted on a robotic arm and equipped with Inertial Measurement Units (IMUs) on the fingertips. These sensors detect the accelerations arisen from contact with external items. In response to a contact, the hand pose and closure are planned for grasping, by executing arm motions with hand closure commands. We generate these motions from human wrist poses acquired from a human maneuvering the SH to grasp an object from a table. We obtained 86% of successful grasps, considering many objects passed to the SH in different manners. We also tested our techniques in preliminary experiments, where the robot moved to autonomously grasp objects from a surface. Results are positive and open interesting perspectives for soft robotic manipulation.

This paper presents methods that facilitate the implementation of dynamic walking on torque-controlled robots in real world experiments. The work uses the Divergent Component of Motion (DCM) for walking trajectory generation and control. The DCM controller is embedded into a whole-body controller (WBC) that produces a full-body walking behavior. While in simulation the combination of DCM and WBC is sufficient for achieving sophisticated walking gaits, during our initial experiments several real-world issues, detailed in this paper, prevented the original control framework from functioning. This work presents the improvements to the original control framework that enabled a breakthrough on the way to achieving torque-based dynamic walking on a real robot.

Visual robot navigation within large-scale, semistructured environments deals with various challenges such as computation intensive path planning algorithms or insufficient knowledge about traversable spaces. Moreover, many state-of-the-art navigation approaches only operate locally instead of gaining a more conceptual understanding of the planning objective. This limits the complexity of tasks a robot can accomplish and makes it harder to deal with uncertainties that are present in the context of real-time robotics applications. In this work, we present Topomap, a framework which simplifies the navigation task by providing a map to the robot which is tailored for path planning use. This novel approach transforms a sparse feature-based map from a visual Simultaneous Localization And Mapping (SLAM) system into a three-dimensional topological map. This is done in two steps. First, we extract occupancy information directly from the noisy sparse point cloud. Then, we create a set of convex free-space clusters, which are the vertices of the topological map. We show that this representation improves the efficiency of global planning, and we provide a complete derivation of our algorithm. Planning experiments on real world datasets demonstrate that we achieve similar performance as RRT* with significantly lower computation times and storage requirements. Finally, we test our algorithm on a mobile robotic platform to prove its advantages.

Nearest-neighbor finding is a major bottleneck for sampling-based motion planning algorithms. The cost of finding nearest neighbors grows with the size of the roadmap, leading to significant slowdowns for problems which require many configurations to find a solution. Prior work has investigated relieving this pressure with quicker computational techniques, such as kd-trees or locality-sensitive hashing. In this work, we investigate an alternative direction for expediting this process based on workspace connectivity. We present an algorithm called Topological Nearest-Neighbor Filtering, which employs a workspace decomposition to select a topologically relevant set of candidate neighbor configurations as a pre-processing step for a nearest-neighbor algorithm. We investigate the application of this filter to several varieties of RRT and demonstrate that the filter improves both nearest-neighbor time and overall planning performance.

In this paper we introduce a novel concept, topological belief space planning (BSP), that uses topological properties of the underlying factor graph representation of future posterior beliefs to direct the search for an optimal solution. This concept deviates from state-of-the-art BSP approaches and is motivated by recent results which indicated, in the context of graph pruning, that topological properties of factor graphs dominantly determine the estimation accuracy. Topological space is also often less dimensional than the embedded state space. In particular, we show how this novel concept can be used in multi-robot belief space planning in high-dimensional state spaces to overcome drawbacks of state-of-the-art approaches: computational intractability of an exhaustive objective evaluation for all candidate path combinations from different robots and dependence on the initial guess in the announced path approach, which can lead to a local minimum of the objective function. We demonstrate our approach in a synthetic simulation.

In this work, we present a novel method for constructing a topological map of biological hotspots in an aquatic environment using a Fast Marching-based Voronoi segmentation. Using this topological map, we develop a closed form solution to the scheduling problem for any single path through the graph. Searching over the space of all paths allows us to compute a maximally informative path that traverses a subset of the hotspots, given some budget. Using a greedy-coverage algorithm we can then compute an informative path. We evaluate our method in a set of simulated trials, both with randomly generated environments and a real-world environment. In these trials, we show that our method produces a topological graph which more accurately captures features in the environment than standard thresholding techniques. Additionally, We show that our method can improve the performance of a greedy-coverage algorithm in the informative path planning problem by guiding it to different informative areas to help it escape from local maxima.

Given a geometric path, the Time-Optimal Path Tracking problem consists in finding the control strategy to traverse the path time-optimally while regulating tracking errors. A simple yet effective approach to this problem is to decompose the controller into two components: (i) a path controller, which modulates the parameterization of the desired path in an online manner, yielding a reference trajectory; and (ii) a tracking controller, which takes the reference trajectory and outputs joint torques for tracking. However, there is one major difficulty: the path controller might not find any feasible reference trajectory that can be tracked by the tracking controller because of torque bounds. In turn, this results in degraded tracking performances. Here, we propose a new path controller that is guaranteed to find feasible reference trajectories by accounting for possible future perturbations. The main technical tool underlying the proposed controller is Reachability Analysis, a new method for analyzing path parameterization problems. Simulations show that the proposed controller outperforms existing methods.

We propose a self-supervised approach for learning representations and robotic behaviors entirely from unlabeled videos recorded from multiple viewpoints, and study how this representation can be used in two robotic imitation settings: imitating object interactions from videos of humans, and imitating human poses. Imitation of human behavior requires a viewpoint-invariant representation that captures the relationships between end-effectors (hands or robot grippers) and the environment, object attributes, and body pose. We train our representations using a triplet loss, where multiple simultaneous viewpoints of the same observation are attracted in the embedding space, while being repelled from temporal neighbors which are often visually similar but functionally different. This signal causes our model to discover attributes that do not change across viewpoint, but do change across time, while ignoring nuisance variables such as occlusions, motion blur, lighting and background. We demonstrate that this representation can be used by a robot to directly mimic human poses without an explicit correspondence, and that it can be used as a reward function within a reinforcement learning algorithm. While representations are learned from an unlabeled collection of task-related videos, robot behaviors such as pouring are learned by watching a single 3rd-person demonstration by a human. Reward functions obtained by following the human demonstrations under the learned representation enable efficient reinforcement learning that is practical for real-world robotic systems. Video results, open-source code and dataset are available at sermanet.github.io/imitate.

This paper presents algorithms for three-dimensional tracking of surgical needles using the stereo endoscopic camera images obtained from the da Vinci® Surgical Robotic System. The proposed method employs Bayesian state estimation, computer vision techniques, and robot kinematics. A virtual needle rendering procedure is implemented to create simulated images of the surgical needle under the da Vinci ® robot endoscope, which makes it possible to measure the similarity between the rendered needle image and the real needle. A particle filter algorithm using the mentioned techniques is then used for tracking the surgical needle. The performance of the tracking is experimentally evaluated using an actual da Vinci® surgical robotic system and quantitatively validated in a ROS/Gazebo simulation thereof.

Wind power is the source of propulsive energy for sailboats and albatrosses. We present the UNAv, an Unmanned Nautical Air-water vehicle, that borrows features from both. It is composed of a glider-type airframe fitted with a vertical wing-sail extending above the center of mass of the system and a vertical surface-piercing hydrofoil keel extending below. The sail and keel are both actuated in pitch about their span-wise axes. Like an albatross, the UNAv is fully streamlined, high lift-to-drag ratio and generates the gravity-cancelling force by means of its airborne wings. Like a sailboat, the UNAv interacts with water and may access the full magnitude of the wind. A trim analysis predicts that a 3.4-meter span, 3 kg system could stay airborne in winds as low as 2.8 m/s (5.5 knots), and travel several times faster than the wind speed. Trim flight requires the ability to fly at extreme low height with the keel immersed in water. For that purpose, a multi-input longitudinal flight controller that leverages fast flap actuation is presented. The flight maneuver is demonstrated experimentally.

In this paper we present an approach for analyzing the inertial effects of changing the wing shape for steering a bat-like robot. Using BaTboT, a robotic platform with massive morphing-wings, we have estimated the generation of pitching and rolling torques, which are directly related to forward and turning maneuvers. Results let us conclude that faster retraction of the wings during the upstroke, and slower extension during the downstroke increase both pitching and rolling torques in about 50% compared to those wingbeats with equal periods for retraction/extension. Also, we determined that the pitch torque generation is proportional to 0.6m1/f, whereas the rolling torque is promotional to 0.1m1/f, being m the mass of the robot and f the flapping frequency of the wings.

This paper demonstrates an autonomous mobile robot that follows a walking user while staying ahead of them. Despite several useful applications for autonomous push-carts, this problem has received much less attention than the easier problem of following from behind. In contrast to previous work, we use multi-modal person detection and a human-motion model that considers obstacles to predict the future path of the user. We implement the system with a modular architecture of obstacle mapper, human tracker, human motion model, robot motion planner and robot motion controller. We report on the performance of the robot in real-world experiments. We believe that approaches to this largely overlooked problem could be useful in real industrial, domestic and entertainment applications in the near future.

To enable robots to perform human-level tasks flexibly in varying conditions, we need a mechanism that allows them to exchange knowledge between themselves for crowd-sourcing the knowledge gap problem. One approach to achieve this is to equip a cloud application with a range of encyclopedic knowledge (i.e. ontologies) and execution logs of different robots performing the same tasks in different environments. In this paper, we show how knowledge exchange between robots can be done using OPENEASE as the cloud application. We equipped OPENEASE with ontologies about the kitchen domain, execution logs of three robots operating in two different kitchens, and semantic descriptions of both environments. By addressing two different use cases, we show that two PR2 robots and one Fetch robot can successfully adapt each other's plan parameters and sub symbolic data to the experiments that they are conducting.

One of the constraints in the design of dry switchable adhesives is the compliance trade-off: compliant structures conform better to surfaces but are limited in strength due to high stored strain energy. In this work we study the effects of bending compliance on the shear adhesion pressures of hybrid electrostatic/gecko-like adhesives of various areas. We reaffirm that normal electrostatic preload increases contact area and show that it is more effective on compliant adhesives. We also show that the gain in contact area can compensate for low shear stiffness and adhesives with high bending compliance outperform stiffer adhesives on substrates with large scale roughness.

We consider the problem of localization in multiagent formations with bearing only measurements, and analyze the fundamental observability properties for dynamic agents. The current well-established approach is based on the socalled rigidity matrix, and its algebraic properties (e.g., its rank and nullspace). This method is typically motivated using first-order derivatives, and shows, among other facts, that the global scale of the formation is not observable. This work shows that current results represent an incomplete view of the problem. In particular, we show that 1) current methods are a particular instantiation of nonlinear observability theory, 2) we can introduce the concept of the dynamic bearing observability matrix from higher order derivatives to study the observability of dynamic formations, and 3) the global scale is, in fact, generally observable when the agents move according to known inputs. We use tools from Riemannian geometry and Lie group theory to tackle, in a general and principled way, the general formulation of the localization problem with states that include both rotations and translations. Finally, we verify our theoretical results by deriving and applying, in both simulations and real experiments on UAVs, a centralized Extended Kalman Filter on Lie groups that is able to estimate the global scale of a moving formation.

We introduce an extension of the Dubins Traveling Salesman Problem with Neighborhoods into the 3D space in which a fixed-wing aerial vehicle is requested to visit a set of target regions while the vehicle motion constraints are satisfied, i.e., the minimum turning radius and maximum climb and dive angles. The primary challenge is to address both the combinatorial optimization part of finding the sequence of target visits and the continuous optimization part of the final trajectory determination. Due to its high complexity, we propose to address both parts of the problem separately by a decoupled approach in which the sequence is determined by a new distance function designed explicitly for the utilized 3D Dubins Airplane model. The final trajectory is then found by a local optimization which improves the solution quality. The proposed approach provides significantly better solutions than using Euclidean distance in the sequencing part of the problem. Moreover, the found solutions are of the competitive quality to the sampling-based algorithm while its computational requirements are about two orders of magnitude lower.

This paper investigates the connection between the kinematics of robots arms and the shortest paths for mobile robots. Lagrange multipliers are used to show that the shortest paths are equivalent to arms in configurations that balance an external force, while applying equal torques and forces at each joint. Analysis of the arm Jacobian yields a further geometric interpretations of optimal paths, constraining the locations of rotation centers and the directions of translations that may occur along optimal paths.

Autonomous driving is a topic in computer vision which has captured a great deal of attention in recent years. One key problem is the detection and state analysis of traffic lights. Even over time, very few datasets for research in this topic have been published and they vary widely in quantity and in quality. To address the complexity of traffic light recognition, we introduce the DriveU**driveU is a joint innovation center of the Daimler AG and the University of Ulm Traffic Light Dataset (DTLD), a large-scale dataset consisting of more than 230,000 annotations. All annotations are hand-labeled according to strict rules and show a high quality. Recordings were made in eleven different cities during different weather conditions. Our dataset exceeds previous traffic light datasets in size, variance, annotation quality and amount of additional sensor data. We prove the extent of our dataset by an extensive comparison with existing traffic light datasets. Miscellaneous dataset criteria are compared, illustrated and statistically analyzed. In the process, metrics to express the quality and variance of datasets are developed and verified. The dataset can be downloaded from http://traffic-light-data.de.

In this paper, we propose a generative model which learns the relationship between language and human action in order to generate a human action sequence given a sentence describing human behavior. The proposed generative model is a generative adversarial network (GAN), which is based on the sequence to sequence (SEQ2SEQ) model. Using the proposed generative network, we can synthesize various actions for a robot or a virtual agent using a text encoder recurrent neural network (RNN) and an action decoder RNN. The proposed generative network is trained from 29,770 pairs of actions and sentence annotations extracted from MSR-Video-to-Text (MSR-VTT), a large-scale video dataset. We demonstrate that the network can generate human-like actions which can be transferred to a Baxter robot, such that the robot performs an action based on a provided sentence. Results show that the proposed generative network correctly models the relationship between language and action and can generate a diverse set of actions from the same sentence.