ICRA 2015 论文列表

In needle-based medical procedures, beveled-tip flexible needles are steered inside soft tissue with the aim of reaching pre-defined target locations. The efficiency of needle-based interventions depends on accurate control of the needle tip. This paper presents a comprehensive mechanics-based model for simulation of planar needle insertion in soft tissue. The proposed model for needle deflection is based on beam theory, works in real-time, and accepts the insertion velocity as an input that can later be used as a control command for needle steering. The model takes into account the effects of tissue deformation, needle-tissue friction, tissue cutting force, and needle bevel angle on needle deflection. Using a robot that inserts a flexible needle into a phantom tissue, various experiments are conducted to separately identify different subsets of the model parameters. The validity of the proposed model is verified by comparing the simulation results to the empirical data. The results demonstrate the accuracy of the proposed model in predicting the needle tip deflection for different insertion velocities.

RAMBLER Robot is designed for researching insect inspired behavioral control algorithms. To evaluate these algorithms, RAMBLER Robot needs autonomous localization without typical sensors like wheel odometers or GPS. The primary objective of this work is to independently, accurately, and robustly recover the path of a moving robotic system with low-cost sensors available off-the-shelf. The computationally efficient power center method of triangulation is compared to a particle filter approach. With three passive indistinguishable landmarks at corners of a small arena, RAMBLER Robot successfully localizes with an RMS error of 2.27cm compared to an overhead camera ground truth.

It is important to understand which view is better recognizing and reconstructing a scene for many robotic applications, especially in a cluttered environment, where objects interact and may occlude one another in all views. In this paper, we introduce a novel, learning-based approach to evaluate scene recognizability from a view based on the quality and quantity of recognized objects, the recognition uncertainty, and the background recognizability, rather than the visibility. Our study shows that increasing visibility does not guarantee better recognizability of objects. The introduced view evaluator can better characterize which view is more useful for the purpose of autonomous object recognition and scene reconstruction. The approach is validated through experiments, and the effects of many factors to scene recognizability are discussed based on the experimental results.

This work proposes a kinematic control method for human-robot handshake motions achieving fast motion synchronization given a preset internal robot handshake motion and a compliance level that reflects the robot's level of passiveness. The proposed method combines a non-linear dynamic system having an attractive limit cycle with an admittance controller and an adaptation mechanism so that interaction forces are minimized and a consensus oscillation is achieved between the engaging participants. The proposed method is validated by experimenting with a KUKA LWR4+ 7dof arm under various scenarios.

This paper presents the development of a new type of robot capable of vertical and directional jumping. The robot uses soft silicone elastomer based pneumatic actuators as legs that accelerate the platform upwards by rapid pressurization. The robot is able to control and adjust the direction of the jumping by altering the timing patterns in which the individual legs are activated.

We developed a ‘hybrid actuator system’ to precisely control the position of a micro-particle at high speed in a microfluidic chip. The hybrid actuator system benefits from two actuator systems by combining an electroosmotic flow generator and pressure-driven flow induced by a piezoelectric actuator-driven syringe pump system. We also implement an image processing algorithm using a high speed vision system and design a supervisory controller for the hybrid actuator system to apply it to a microfluidic system. The proposed system can be used to overcome current limitations of conventional pump systems for microfluidic devices.

We present an algorithm that seeks to find a set of diverse, short paths through a roadmap graph. The usefulness of a such a set is illustrated in robotic motion planning and routing applications wherein a precomputed roadmap of the environment is partially invalidated by some change, for example, relocation of obstacles or reconfiguration of the robot. Our algorithm employs the heuristic that nearby configurations are likely to be invalidated by the same change. To find diverse short paths, the algorithm finds the shortest detour avoiding a collection of balls imposed on the graph as simulated obstacles. Different collections yield different short paths. Paths may then be checked for validity as a cheap alternative to checking or reconstructing the entire roadmap. We describe a formal definition of path set diversity and several measures on which to evaluate our algorithm. We compare the speed and quality of our heuristic algorithm's results against an exact algorithm that computes the optimally shortest set of paths on the roadmap having a minimum diversity. We show that, with tolerable loss in shortness, we produce equally diverse path sets orders of magnitude more quickly.

We formulate and study the comprehensive multi-root (CMR) planning problem, in which feasible paths are desired between multiple regions. We propose two primary contributions which allow us to extend state-of-the-art sampling-based planners. First, we propose the notion of vertex coloring as a compact representation of the CMR objective on graphs. Second, we propose a method for deferring edge evaluations which do not advance our objective, by way of a simple criterion over these vertex colorings. The resulting approach can be applied to any CMR-agnostic graph-based planner which evaluates a sequence of edges. We prove that the theoretical performance of the colored algorithm is always strictly better than (or equal to) that of the corresponding uncolored version. We then apply the approach to the Probabalistic RoadMap (PRM) algorithm; the resulting Colored Probabalistic RoadMap (cPRM) is illustrated on 2D and 7D CMR problems.

Pivoting is the rotation of an object between two fingers using gravity and inertial forces to impart angular momentum. We present an analysis of the mechanics of pivoting and a framework for planning and execution. Extrinsic dexterity was defined by Chavan-Dafle et al. [1] as the use of external forces, such as gravity and inertial forces in post grasp manipulation. We analyze one such regrasp termed “pivoting” by Rao et al. [2]. We find a grasp and arm trajectory which can rotate an object between stable poses, if any. We demonstrate an implementation of pivoting with an ABB industrial arm and a two fingered gripper.

We propose a novel algorithm for efficient exploration with a single agent in unknown environments, populated with static obstacles. Every next position is computed by employing environment inference on top of path planning. The path planning process that ensures obstacle avoidance uses a modified version of the A* algorithm [1] and the inference is performed by direct sensing and by predicting values at yet-not-visited positions using Gaussian processes. We have validated our algorithm with densely measured data of an indoor magnetic field with significant spatial variations in different obstacle setups. Additionally, it is shown that it outperforms a previously developed algorithm [2], for obstacle-free scenarios, and the random movement of the agent.

Learning motion tasks in a real environment with deformable objects requires not only a Reinforcement Learning (RL) algorithm, but also a good motion characterization, a preferably compliant robot controller, and an agent giving feedback for the rewards/costs in the RL algorithm. In this paper, we unify all these parts in a simple but effective way to properly learn safety-critical robotic tasks such as wrapping a scarf around the neck (so far, of a mannequin).

This paper presents a universally applicable graph-based framework for the navigation of warehouse robots equipped with only monocular cameras. We strongly advocate the use of relative pose information stored in a topological map, rather than a globally consistent metric representation of the environment. We show how multiple traversals of adjacent workspaces can be naturally “stitched” together in the course of a typical warehouse picking and shelving schedule to create a network of reusable paths in which the robot can efficiently localise and plan new routes. This allows us to command the robot to return to any of the previously visited locations not necessarily through the same route that we taught it. Unlike state-of-the-art teach and repeat systems using stereo vision, our approach exploits the strongly planar nature of the data obtained from a downward-facing camera, and creates odometric constraints by tracking the perceived texture of the floor and computing a simple homography. To demonstrate the robustness of our system, we validate our approach on datasets collected over a week-long period within a challenging and representative environment in the form of a warehouse shelving area.

Small Manufacturing Entities (SMEs) have not incorporated robotic automation as readily as large companies due to rapidly changing product lines, complex and dexterous tasks, and the high cost of start-up. While recent low-cost robots such as the Universal Robots UR5 and Rethink Robotics Baxter are more economical and feature improved programming interfaces, based on our discussions with manufacturers further incorporation of robots into the manufacturing work flow is limited by the ability of these systems to generalize across tasks and handle environmental variation. Our goal is to create a system designed for small manufacturers that contains a set of capabilities useful for a wide range of tasks, is both powerful and easy to use, allows for perceptually grounded actions, and is able to accumulate, abstract, and reuse plans that have been taught. We present an extension to Behavior Trees that allows for representing the system capabilities of a robot as a set of generalizable operations that are exposed to an end-user for creating task plans. We implement this framework in CoSTAR, the Collaborative System for Task Automation and Recognition, and demonstrate its effectiveness with two case studies. We first perform a complex tool-based object manipulation task in a laboratory setting. We then show the deployment of our system in an SME where we automate a machine tending task that was not possible with current off the shelf robots.

In this paper a novel teleoperation framework for aerial robots that physically interact with the environment is presented. This framework allows to teleoperate the robot both in contact-free flight and in physical contact with the environment in order, e.g., to apply desired forces on objects of the environment. The framework is build upon an impedance-like indirect interaction force controller that allows to use standard underactuated aerial robots as force effectors. Haptic feedback from the master side enables the user to feel the contact forces exerted by the robot. An automatic potential field-based slowing-down policy is used by the robot to ensure a smooth transition between the contact-free motion phase and the force interaction phase. The effectiveness of the approach has been shown in extensive human-in-the-loop simulations including remote pressing of buttons on a surface and pushing a cart until it touches a wall.

In this paper, we present an object-centric, fixed-dimensional 3D shape representation for robust matching of partially observed object shapes, which is an important component for object categorization from 3D data. A main problem when working with RGB-D data from stereo, Kinect, or laser sensors is that the 3D information is typically quite noisy. For that reason, we accumulate shape information over time and register it in a common reference frame. Matching the resulting shapes requires a strategy for dealing with partial observations. We therefore investigate several distance functions and kernels that implement different such strategies and compare their matching performance in quantitative experiments. We show that the resulting representation achieves good results for a large variety of vision tasks, such as multi-class classification, person orientation estimation, and articulated body pose estimation, where robust 3D shape matching is essential.

We propose a semantic scene understanding system that is suitable for real robotic operations. The system solves different tasks (semantic segmentation and object detections) in an opportunistic and distributed fashion but still allows communication between modules to improve their respective performances. We propose the use of the semantic space to improve specific out-of-the-box object detectors and an update model to take the evidence from different detection into account in the semantic segmentation process. Our proposal is evaluated with the KITTI dataset, on the object detection benchmark and on five different sequences manually annotated for the semantic segmentation task, demonstrating the efficacy of our approach.

This paper discusses a new control approach for robotic-assisted tele-echography where contact dynamics are driven by a 3D time-of-flight camera and a force sensor. Based on 3D camera and force data, contact stiffness is anticipated, allowing adaptation before contact. This approach enables to slow down robot dynamics before contact which is adequate for tele-echographic tasks. Additionally, free space inertial forces can also be eliminated from the control loop. In contact, environment stiffness is estimated online using only force data. The control architecture features a dynamically consistent hierarchical approach, where an explicit Cartesian force control arises as the primary task while orientation control is designed in the null space. Control performance is boosted by Kalman active observers (AOBs). Experiments with a 7-DOF WAM™ validate the architecture, where an ultrasound probe is teleoperated in 6 Cartesian dimensions, executing typical echographic tasks.

In this paper we propose a novel algorithm for the dynamic routing of a group of Autonomous Guided Vehicles (AGVs) used for transporting goods in automatic warehouses. Our strategy allows to improve the efficiency of a fleet of AGVs in terms of delivery time and its computational burden is sufficiently small to be embedded in standard industrial traffic management system. The algorithm is validated in a small-scale automatic warehouse with real AGVs.

Population-based learning techniques have been proven to be effective in dealing with noise in numerical benchmark functions and are thus promising tools for the high-dimensional optimization of controllers for multiple robots with limited sensing capabilities, which have inherently noisy performance evaluations. In this article, we apply a statistical technique called Optimal Computing Budget Allocation to improve the performance of Particle Swarm Optimization in the presence of noise for a multi-robot obstacle avoidance benchmark task. We present a new distributed PSO OCBA algorithm suitable for resource-constrained mobile robots due to its low requirements in terms of memory and limited local communication. Our results from simulation show that PSO OCBA outperforms other techniques for dealing with noise, achieving a more consistent progress and a better estimate of the ground-truth performance of candidate solutions. We then validate our simulations with real robot experiments where we compare the controller learned with our proposed algorithm to a potential field controller for obstacle avoidance in a cluttered environment. We show that they both achieve a high performance through different avoidance behaviors.