ICRA 2018 论文列表

Ophthalmic anesthesia is an important preparation for eye surgery. The conventional practice is performed blind in a cadaver under the supervision of an experienced surgeon. This paper introduces a needle tip tracking system for ophthalmic anesthesia training without major modification of an anesthesia needle. The study presents a prototyped system to track a magnetized needle tip using Hall-effect sensor array. The orbital structure model was embedded with Hall-effect sensors after considering the sensing workspace and ophthalmic anesthesia pathway. The extended Kalman filter was used to calculate needle tip position. A commercial robotic manipulator was used to model the characteristics of sensor and accuracy of the developed system. A prototype can detect needle tip position with a root-mean-square deviation around 1.80 mm. As a result, the system is capable of providing needle tip positions for training purposes.

We present a data-driven, bottom-up, deep learning approach to robotic grasping of unknown objects using Deep Convolutional Neural Networks (DCNNs). The approach uses depth images of the scene as its sole input for synthesis of a single-grasp solution during execution, adequately portraying the robot's visual perception during exploration of a scene. The training input consists of precomputed high-quality grasps, generated by analytical grasp planners, accompanied with rendered depth images of the training objects. In contrast to previous work on applying deep learning techniques to robotic grasping, our approach is able to handle full end-effector poses and therefore approach directions other than the view direction of the camera. Furthermore, the approach is not limited to a certain grasping setup (e. g. parallel jaw gripper) by design. We evaluate the method regarding its force-closure performance in simulation using the KIT and YCB object model datasets as well as a big data grasping database. We demonstrate the performance of our approach in qualitative grasping experiments on the humanoid robot ARMAR-III.

This paper presents an online data collection method that captures human intuition about what grasp types are preferred for different fundamental object shapes and sizes. Survey questions are based on an adopted taxonomy that combines grasp pre-shape, approach, wrist orientation, object shape, orientation and size which covers a large swathe of common grasps. For example, the survey identifies at what object height or width dimension (normalized by robot hand size) the human prefers to use a two finger precision grasp versus a three-finger power grasp. This information is represented as a confidence-interval based polytope in the object shape space. The result is a database that can be used to quickly find potential pre-grasps that are likely to work, given an estimate of the object shape and size.

In this paper we propose a grasp strategy which exploits environmental contact for grasping domestic flat objects placed or hinged on support surfaces. The proposed grasp strategy considers the non-convex geometry of the object-surface combination, as this appears in objects like plates on tables or handles on cupboards. Following the fact that state-of-the-art grasp planners fail to produce candidate grasps for flat objects due to the environmental constraint of the support surface, this work utilizes compliant interaction of the hand with the support surface, inspired by human grasp strategies.

In this paper, we present the design of a novel, hybrid multipurpose robotic platform equipped with a pair of graspers to synergize grasping, manipulation, and locomotion. The multipurpose grasper consists of two underactuated fingers with an active gripping surface, which passively conforms to an object while grasping. Each finger has a spring loaded synchronous belt drive which functions as the active gripping surface. The grasper is capable of handling a range of objects with irregular geometry and size. Two such underactuated graspers are connected through a serial kinematic chain and this provides the platform both manipulation and locomotion capability. Graspers act as “legs” or “wheels” of the robot during locomotion and can easily adapt to terrain variations. Fewer number of actuators, simple and scalable kinematic structure, and computationally efficient control are some of the main features of the design. Design details and kinematic analysis are presented. Experiments were conducted on a prototype robot to demonstrate multiple modes of operation.

Using a rehabilitation robot to activate motion intention and reflex response simultaneously is an effective approach to aiding recovery from paralysis caused by neurological disorders. Mechanical motions supported by conventional robots are, however, not enough to activate reflex. In this paper, we propose a grasp-training robot that can stimulate the grasp reflex of a paralyzed hand by pushing the hand onto an elastic bar while supporting the grasping movements. In addition to this feature, we discuss the robot design in relation to its usability and wearability for ease of use in clinical practice. Experimental results obtained from healthy subjects show that the proposed robot can support grasping in a way similar to the traditional range-of-motion exercise used by therapists for grasp rehabilitation. Combining this appropriate grasping-motion support and the mechanism for pushing the hand onto an elastic bar succeeds in activating the grasp reflex of a completely paralyzed patient in a clinical test that involves monitoring electromyography signals from the paralyzed hand.

Grasping a moving target has been investigated extensively for fixed-base manipulator. However, such a task becomes much more challenging when the manipulator is free flying in the air with an UAV. Towards moving target grasping, this paper presents an aerial manipulator system composed of a hex-rotor and a 7-DoF (Degree of Freedom) manipulator. An independent control structure is used in the aerial manipulator control system, i.e., the hex-rotor and the manipulator are controlled separately. In the hex-rotor's controller, the system CoM (Center of Mass) offset motion is used to compensate disturbance of the robotic arm. In the manipulator's controller, the relative kinematics between the target and the aerial vehicle is taken into consideration to grasp the target. At last aerial grasping experiments are conducted to validate the feasibility of the proposed control scheme and the reliability of our aerial manipulator system.

In this paper we propose a new approach to evaluate grasps that accounts for both the kinematic structure of the robot and the noise at its joints. Our starting observation is that with a redundant robot the same grasp can be implemented with different arm configurations, and these may display significant differences in terms of robustness to disturbances. Consequently, the grasp quality metric is seen as a random variable depending on the arm configuration. Starting from a first order approximation for the error, we introduce the high probability force closure region as a tool to evaluate the local robustness of an arm configuration, and we then introduce a new metric Qarm to rank different configurations according to the robustness to noise. By combining this method in an offline/online framework, we demonstrate through large scale simulations that this approach successfully captures aspects that were neglected in former literature regarding grasp evaluation, and can successfully be integrated into future grasp planners.

In near future, robots are envisioned to work alongside humans in unstructured professional and domestic environments. In such setups, collaborative manipulation is a fundamental skill that allows manipulation of heavy loads by load sharing between agents. Grasp planning plays a pivotal role for load sharing but it has not received attention in the literature. This work proposes a grasp analysis approach for collaborative manipulation that allows load sharing by minimizing exerted grasp wrenches in a task specific way. The manipulation task is defined as expected external wrenches acting on the target object. The analysis approach is demonstrated in a two-agent decentralized set-up with unknown objects. After the first agent has grasped the target, the second agent observes the first agent's grasp location and plans its own grasp according to optimal load sharing. The method was verified in a human robot collaborative lifting task. Experiments with multiple objects show that the proposed method results in optimal load sharing despite limited information and partial observability.

Planning smooth trajectories is important for the safe, efficient and comfortable operation of mobile robots, such as wheeled robots moving in crowded environments or cars moving at high speed. Asymptotically optimal sampling-based motion planners can be used to generate such trajectories. However, to achieve the necessary efficiency for the realtime operation of robots, one often uses their initial feasible trajectories or the trajectories of non-optimal motion planners instead, typically after a post-smoothing step. We propose a gradient-informed post-smoothing algorithm, called GRIPS, that deforms given trajectories by locally optimizing the placement of vertices while satisfying the system's kinodynamic constraints. We show experimentally that GRIPS typically produces trajectories of significantly smaller length and higher smoothness than several existing post-smoothing algorithms.

We develop a general control framework where a low-level optimizer is built into the robot dynamics. This optimizer together with the robot constitute a goal directed dynamical system, controlled on a higher level. The high level command is a cost function. It can encode desired accelerations, end-effector poses, center of pressure, and other intuitive features that have been studied before. Unlike the currently popular quadratic programming framework, which comes with performance guarantees at the expense of modeling flexibility, the optimization problem we solve at each time step is non-convex and non-smooth. Nevertheless, by exploiting the unique properties of the soft-constraint physics model we have recently developed, we are able to design an efficient solver for goal directed dynamics. It is only two times slower than the forward dynamics solver, and is much faster than real time. The simulation results reveal that complex movements can be generated via greedy optimization of simple costs. This new computational infrastructure can facilitate teleoperation, feature-based control, deep learning of control policies, and trajectory optimization. It will become a standard feature in future releases of the MuJoCo simulator.

Robots fabricated by soft materials can provide higher flexibility and thus better safety while interacting with natural objects with low stiffness such as food and human beings. However, as many more degrees of freedom are introduced, the motion simulation of a soft robot becomes cumbersome, especially when large deformations are presented. Moreover, when the actuation is defined by geometry variation, it is not easy to obtain the exact loads and material properties to be used in the conventional methods of deformation simulation. In this paper, we present a direct approach to take the geometric actuation as input and compute the deformed shape of soft robots by numerical optimization using a geometry-based algorithm. By a simple calibration, the properties of multiple materials can be modeled geometrically in the framework. Numerical and experimental tests have been conducted to demonstrate the performance of our approach on both cable-driven and pneumatic actuators in soft robotics.

Accuracy of robots is an important facet in an industrial setting. In this paper, we present a novel kinematic calibration methodology. Traditional calibration techniques require an external metrology device. Unlike those, the presented product here is highly practical in that it does not require a metrology device. The kinematic calibration model is formulated by making use of the robot itself as a metrology device to measure the geometry of a known artifact. The optimal parameters/characteristics of the model are identified using a Particle Swarm Optimization (PSO) technique. Our experimental results show that this new approach provides results comparable to those generated using spatial information provided by a Coordinate Measurement Machine (CMM). Using this new approach (GageCAL), the Yaskawa Motoman “MHS-Hi” robot is calibrated. Our experimental testing also indicates that this methodology can be extended to a wide variety of anthropomorphic robots.

This paper explores the problem of autonomous, in-hand regrasping-the problem of moving from an initial grasp on an object to a desired grasp using the dexterity of a robot's fingers. We propose a planner for this problem which alternates between finger gaiting, and in-grasp manipulation. Finger gaiting enables the robot to move a single finger to a new contact location on the object, while the remaining fingers stably hold the object. In-grasp manipulation moves the object to a new pose relative to the robot's palm, while maintaining the contact locations between the hand and object. Given the object's geometry (as a mesh), the hand's kinematic structure, and the initial and desired grasps, we plan a sequence of finger gaits and object reposing actions to reach the desired grasp without dropping the object. We propose an optimization based approach and report in-hand regrasping plans for 5 objects over 5 in-hand regrasp goals each. The plans generated by our planner are collision free and guarantee kinematic feasibility.

This paper deals with an OCT (optical coherence tomography) geometric calibration method. OCT medical imaging system has received a growing interest during the last two decades. In medical purposes, OCT images are generally called optical biopsies which allows in-vivo investigation almost similar to a histopathological study. The physician can rely on the OCT images to establish a rapid and direct diagnosis. But the OCT images formation suffered numerous distortions due in particular to the optical path, from the source to the viewed sample passing through the two reflecting mirrors and a scan objective. The obtained optical biopsies include several spectral and geometric distortions. The proposed calibration model aims to compensate the geometrical ones. More precisely, two models were developed allowing the correction of both 2D images (B-Scan slices) and 3D images (volume). These models were experimentally validated (in both artificial and biological samples) using a spectral domain OCT system. It has demonstrated a significant enhancement of the OCT images accuracy.

Highly Autonomous Driving (HAD) systems rely on deep neural networks for the visual perception of the driving environment. Such networks are train on large manually annotated databases. In this work, a semi-parametric approach to one-shot learning is proposed, with the aim of bypassing the manual annotation step required for training perceptions systems used in autonomous driving. The proposed generative framework, coined Generative One-Shot Learning (GOL), takes as input single one-shot objects, or generic patterns, and a small set of so-called regularization samples used to drive the generative process. New synthetic data is generated as Pareto optimal solutions from one-shot objects using a set of generalization functions built into a generalization generator. GOL has been evaluated on environment perception challenges encountered in autonomous vision.

Conventional approaches of robotic writing of Chinese character strokes often suffer from limited font generation methods, and thus the writing results often lack of diversity. This has seriously restricted the high quality writing ability of robots. This paper proposes a generative adversarial nets-based calligraphic robotic framework, which enables a robot to learn writing fundamental Chinese strokes with rich diversity and good originality. In particular, the framework considers the learning process of robotic writing as an adversarial procedure which is implemented by three interactive modules including a stroke generation module, a stroke discriminative module and a training module. Noting that the stroke generative module included in the conventional generative adversarial nets cannot solve the non-differentiable problem, the policy gradient commonly used in reinforcement learning is thus adapted in this work to train the generative module by regarding the outputs from the discriminative module as rewards. Experimental results demonstrate that the proposed framework allows a calligraphic robot to successfully write fundamental Chinese strokes with good quality in various styles. The experiment also suggests the proposed approach can achieve human-level stroke writing quality without the requirement of a performance evaluation system. This approach therefore significantly boosts the robotic autonomous creation ability.

This paper presents a new method to generate vibration free rest-to-rest (RTR) trajectories for configuration dependent dynamic systems, such as robots, cranes or machine tools. The new method named 3-Segmented Input Shaping is based on a combination of the widely known Input Shaping method and a new trajectory segmentation strategy for piece wise shaping of the trajectory. The new segmentation strategy facilitates the capability of accounting for variations in system dynamics during motion by shaping acceleration and deceleration profiles with individual frequencies. In this paper the new segmentation strategy is used in combination with the bang-coast-bang (BCB) trajectory. The generated trajectories are described in closed form, hence requires no optimization and thereby provides strong computational performance. The new method is verified by numerical simulations and detailed analysis and shows great potential in vibration-free RTR trajectory generation for systems with configuration dependent dynamics.

Human-humanoid collaborative tasks require that the robot take into account the goals of the task, interaction forces with the human, and its own balance. We present a formulation for a real-time humanoid controller which allows the robot to keep itself balanced, while also assisting the human in achieving their shared objectives. We achieve this with a multi-robot quadratic program controller, which solves for human dynamics reconstruction and optimal robot controls in a single optimization problem. Our experiments on a simulated robot platform demonstrate the ability to generate interaction motions and forces that are similar to what a human collaborator would produce.