IROS 2017 论文列表

We present further computational tools and control results in the framework of human-robot coexistence and collaboration. A GPU parallel processing algorithm is introduced for real-time monitoring of dynamic distances between a robot and generic obstacles moving in its environment, taking advantage of the handling of RGB-D data directly in the depth space of the sensor. Combined with the use of model-based residual signals, this approach allows efficient detection of contact points on the robot with simultaneous estimation of the exchanged contact forces. When the robot is kinematically redundant for the original task and undergoes a physical contact, a control scheme accommodates collaboration trying to preserve task execution, or reacts by abandoning the task if the estimated contact forces exceed some safety threshold. Experimental results are reported for a KUKA LWR.

Automated assembly, especially peg-in-hole insertion, is a common task in manufacturing. In particular, the high-precision assembly is achieved by high-precision manipulator and sensing system. However, uncertainty and various parts for assembly are still challenges for robotic assembly, especially for low-precision robot and sensors. It is noteworthy that human can implement assembly tasks although the precision of the arm and hand is not comparable with a common industrial robot, in which process compliance is the key characteristic of their motion. In this paper, we present a human-inspired compliant strategy for peg-in-hole assembly task using the environmental constraint and coarse force information. In the proposed strategy, a constraint region is designed for motion planning and utilized for eliminating the uncertainty of the initial positioning error of the peg. Force sensor is applied to sense the contact force of which the direction is used to adjust the movement of the peg. Therefore, high-precision sensor is not necessarily required. Inspired by human compliant assembly, a from coarse to fine adjustment strategy is executed. The contribution of our strategy is that high precision assembly task can be solved by low precision system. The constraint region and force guided directional adjustment have increased the robustness of the system. The strategy is carried out in simulation for round peg-in-hole assembly task. The experimental results show that the assembly task can be successfully completed and demonstrate the effectiveness of our strategy.

This paper addresses the problem of mixed initiative, shared control for master-slave grasping and manipulation. We propose a novel system, in which an autonomous agent assists a human in teleoperating a remote slave arm/gripper, using a haptic master device. Our system is designed to exploit the human operator's expertise in selecting stable grasps (still an open research topic in autonomous robotics). Meanwhile, a-priori knowledge of: i) the slave robot kinematics, and ii) the desired post-grasp manipulative trajectory, are fed to an autonomous agent which transmits force cues to the human, to encourage maximally manipulable grasp pose selections. Specifically, the autonomous agent provides force cues to the human, during the reach-to-grasp phase, which encourage the human to select grasp poses which maximise manipulation capability during the post-grasp object manipulation phase. We introduce a task-oriented velocity manipulability cost function (TOV), which is used to identify the maximum kinematic capability of a manipulator during post-grasp motions, and feed this back as force cues to the human during the pre-grasp phase. We show that grasps which minimise TOV result in significantly reduced control effort of the manipulator, compared to other feasible grasps. We demonstrate the effectiveness of our approach by experiments with both real and simulated robots.

This paper proposes a new algorithm for human motion estimation using inertial measurement unit (IMU) measurements. We model the joints by matrix Lie groups, namely the special orthogonal groups SO(2) and SO(3), representing rotations in 2D and 3D space, respectively. The state space is defined by the Cartesian product of the rotation groups and their velocities and accelerations, given a kinematic model of the articulated body. In order to estimate the state, we propose the Lie Group Extended Kalman Filter (LG-EKF), thus explicitly accounting for the non-Euclidean geometry of the state space, and we derive the LG-EKF recursion for articulated motion estimation based on IMU measurements. The performance of the proposed algorithm is compared to the EKF based on Euler angle parametrization in both simulation and real-world experiments. The results show that for motion near gimbal lock regions, which is common for shoulder movement, the proposed filter is a significant improvement over the Euler angles EKF.

The human forearm is composed of two long, thin bones called the radius and the ulna, and rotates using two axle joints. We aimed to develop a forearm based on the body proportion, weight ratio, muscle arrangement, and joint performance of the human body in order to bring out its benefits. For this, we need to miniaturize the muscle modules. To approach this task, we arranged two muscle motors inside one muscle module, and used the space effectively by utilizing common parts. In addition, we enabled the muscle module to also be used as the bone structure. Moreover, we used miniature motors and developed a way to dissipate the motor heat to the bone structure. Through these approaches, we succeeded in developing a forearm with a radioulnar joint based on the body proportion, weight ratio, muscle arrangement, and joint performance of the human body, while keeping maintainability and reliability. Also, we performed some motions such as soldering, opening a book, turning a screw, and badminton swinging using the benefits of the radioulnar structure, which have not been discussed before, and verified that Kengoro can realize skillful motions using the radioulnar joint like a human.

This paper deals with the problem of using the tactile feedback generated by a robotic skin for discriminating a human hand touch from a generic contact. Humans understand collaboration intentions through different sensing modalities such as vision, hearing and touch. Among them, a physical interaction is mainly used for demonstrating or correcting a kind of motion and is usually started by touching with the hands the other human body. Until recently, it was difficult to perform the same in human-robot cooperation due to the lack of large-scale tactile systems functionally similar to a human skin. Our approach consists in transforming measurements of sensors distributed on the robot body into a convenient 2D representation of the contact shape, i.e., a contact image, then applying image classification techniques in order to discriminate a human touch from unexpected collisions. Experiments have been performed on a robotic skin composed of 768 pressure sensors integrated on a Baxter robot forearm. More than 1800 contact images have been generated from 43 different persons for training and testing two machine learning algorithms: Bag of Visual Words and Convolutional Neural Networks. The experimental results show that both approaches are valid, obtaining a classification accuracy higher than 96%.

Some stroke patients are affected by hemiplegia, which is the complete paralysis of half of the body, and need to train in upper limb motions with high frequency to improve their quality of life. Therefore, the authors developed a portable robot with a main processor, actuators, motor drivers, a force sensor, and a touch display to assist such high-frequency training at home. To recover the motor function of the upper limbs by voluntary movement, this robot instructs the joint motion based on the force input. Therefore, this study proposes a force control algorithm for the omni-wheel mechanism and a method that estimates the odometry of the omni-wheel mechanism based on the circuit equation of the DC motor without rotary encoders. Moreover, a method that estimates integrated electromyograph patterns based on the force and movement of the robot using a multiple linear regression model is proposed for the evaluation of co-operative movement. The effectiveness of these proposed methods were confirmed by evaluation experiments.

Human can not only support their body during standing or walking, but also support them by hand, so that they can dangle a bar and others. But most humanoid robots support their body only in the foot and they use their hand just to manipulate objects because their hands are too weak to support their body. Strong hands are supposed to enable humanoid robots to act in much broader scene. Therefore, we developed new life-size five-fingered hand that can support the body of life-size humanoid robot. It is tendon-driven and underactuated hand and actuators in forearms produce large gripping force. This hand has flexible joints using machined springs, which can be designed integrally with the attachment. Thus, it has both structural strength and impact resistance in spite of small size. As other characteristics, this hand has force sensors to measure external force and the fingers can be flexed along objects though the number of actuators to flex fingers is less than that of fingers. We installed the developed hand on musculoskeletal humanoid “Kengoro” and achieved two self-weight supporting motions: push-up motion and dangling motion.

Optical Tweezers are considered one of the most suitable techniques for biological tasks, however the lack of automation make this technology less accessible. We present here a new 3D force sensing method with high bandwidth (up to 10Khz) which can allow implementing complex robotic approaches. Proposed technique uses high speed image tracking with nano-metric resolution in 3 directions. Its capabilities are demonstrated in a teleoperated 3D manipulation scenario with a haptic user interface, where naive users performed direct in vitro haptic exploration of isolated Red Blood Cells inside a Petri dish.

The traditional programming method for industrial robots is teaching by demonstration with a joystick or teaching pendant. However, high precision welding, machining, painting or thermal spraying, and other tasks require pre-computed trajectories, since an operator would not be able to accurately guide the manipulator and achieve suitable results. Giving the CAD model of a 3-D freeform surface, and the operating parameters of the task, this paper describes the methodology and steps involved in off-line programming for automatic trajectory generation. The computed waypoints comprise position, orientation, manipulator's end-effector velocities, and accelerations. This work describes the software platform employed for simulation, surface sampling, the use of radial basis function (RBF) for surface modelling, and intersection of surfaces (marching method) to generate paths. The case study for the developed method is the EMMA project∗, a robotic system for in situ high velocity oxy-fuel coating (HVOF) of hydraulic turbine blades. The results highlight the efficacy of the method and the impact of the solution for similar problems.

In the field of loop closure detection, the most conventional approach is based on the Bag-of-Visual-Words (BoVW) image representation. Although well-established, this model rejects the spatial information regarding the local feature points' layout and performs the associations based only on their similarities. In this paper we propose a novel BoVW-based technique which additionally incorporates the operational environment's structure into the description, treating bunches of visual words with similar optical flow measurements as single similarity votes. The presented experimental results prove that our method offers superior loop closure detection accuracy while still ensuring real-time performance, even in the case of a low power consuming mobile device.

This paper introduces a generic framework for field surveillance using consumer rotorcrafts and ground vehicles. Building such an autonomous system comes with two key challenges in persistent perception and obstacle avoidance. We begin with explaining two core algorithms to solve the challenges: an auto-landing algorithm that enables a quadrotor to land on a moving ground vehicle at a speed of 6.00 m/s, and an obstacle avoidance algorithm that ensures the safety of the quadrotor during searching process. On the basis of these algorithms, the architecture and infrastructure of Hawkeye framework are presented as well. Hawkeye is designed to be a generic platform with extensibility that allows integration of other domain applications. We demonstrate the potential of Hawkeye framework in a simulated agriculture monitoring mission and report its performance at the end of the paper.

Long-short term memory networks(LSTM) models have shown considerable performance on variety of problems dealing with sequential data. In this paper, we propose a variant of Long-Term Recurrent Convolutional Network(LRCN) to detect road intersection. We call this network as IntersectNet. We pose road intersection detection as binary classification task over sequence of frames. The model combines deep hierarchical visual feature extractor with recurrent sequence model. The model is end to end trainable with capability of capturing the temporal dynamics of the system. We exploit this capability to identify road intersection in a sequence of temporally consistent images. The model has been rigorously trained and tested on various different datasets. We think that our findings could be useful to model behavior of autonomous agent in the real-world.

Handshakes are common greetings, and humans therefore have strong priors of what a handshake should feel like. This makes it challenging to create compelling and realistic human-robot handshakes, necessitating the consideration of human haptic perception in the design of robot hands. At its most basic level, haptic perception is encoded by contact points and contact pressure distributions on the skin. This motivates our work on measuring the contact area and contact pressure in human handshaking interactions. We present two benchmarking experiments in this regard, measuring the contact locations in human-human/human-robot handshaking and the contact pressure distribution for handshakes with a sensorized palm. We present results from human studies with the benchmarking experiments, providing a baseline for comparison with robot hands as well as presenting new insights into human handshaking. We also show initial work in using these results for the evaluation of robot hands, and progressing towards iterative design of robot hands optimized for social hand interactions.

In this paper, the Hardware Embedded Reduced Intricacy (HERI) Hand, which is a novel tendon driven three-finger under-actuated hand demonstrating balanced dexterous finger manipulation and powerful grasping of common objects is presented. The third finger of HERI Hand is asymmetrically designed in terms of dimensions to emulate the functionality for combining middle finger, ring finger and little finger of a human hand. HERI Hand is equipped with three actuators devoted to the actuation of the flexion of the index finger and thumb, the flexion of the third finger and finally the thumb abduction and adduction motion with the latest drive having no interference with other transmissions. The proposed hand is capable of realizing delicate finger manipulation such as opening a lidded cup, which is super suitable to accomplish in such configuration. At the same time the hand demonstrates high grasping strength capacity thanks to the actuation sizing permitted by the under-actuated configuration. The hand is also equipped with tactile/pressure sensors distributed in the phalanxes, which leaves open the possibility of potential applications for sophisticated finger manipulations taking into account the phalanxes contact forces. Three different sets of experiments were carried out to demonstrate the performance of HERI Hand and validated its functionality.

In this paper, we propose a novel approach to integrating inertial sensor data into a pose-graph free dense mapping algorithm that we call GravityFusion. A range of dense mapping algorithms have recently been proposed, though few integrate inertial sensing. We build on ElasticFusion, a particularly elegant approach that fuses color and depth information directly into small surface patches called surfels. Traditional inertial integration happens at the level of camera motion, however, a pose graph is not available here. Instead, we present a novel approach that incorporates the gravity measurements directly into the map: Each surfel is annotated by a gravity measurement, and that measurement is updated with each new observation of the surfel. We use mesh deformation, the same mechanism used for loop closure in ElasticFusion, to enforce a consistent gravity direction among all the surfels. This eliminates drift in two degrees of freedom, avoiding the typical curving of maps that are particularly pronounced in long hallways, as we qualitatively show in the experimental evaluation.

This paper presents an analysis of the design and operation of a non-articulated robotic hand, inspired by the multilayered structure. The hand is made of a non-stretchable thin film, the two ends of which are attached to a soft substrate, forming an enclosed wrinkled shape. When the substrate is elongated, the morphology of the hand changes varied, resulting in a “wrapping” form around the grasped object. This design is applicable to a wide range of objects that can be grasped by the robotic hand. The smallest and biggest sizes of the grasped object are determined by the morphological computation of the film during design of the hand. This design also reduces the uncertainty of position between the hand and the object; i.e., stable grasping can be maintained as long as the object fits within the enclosed form of the hand. A mechanism controlling the morphology of the film using only one actuator, and for grasping food products such as fruits, is proposed. Preliminary results show the potential of the hand's morphology in the design of soft robotic mechanisms.

The growing demand in industry for robots capable of performing a variety of tasks requires an increased capability in robotic grasping. Humans are adept at interacting with novel objects, a skill attributed primarily to tactile feedback in the form of exteroception and proprioception. This paper presents a novel way to incorporate exteroception and proprioception into grasp stability assessment: by using convolutional neural networks. This method improves upon the results of a unsupervised feature learning approach that used similar tactile feedback. 1000 different grasps on 100 objects were used to train and test the network. The network achieved an overall accuracy of 88.4% while predicting the failure class with an accuracy of 92.7%.

This paper deals with the problem of purposefully failing or yielding an object by a robotic gripper. We propose a grasp quality measure fabricated for robotic harvesting in which picking a crop from its stem is desired. The proposed metric characterizes a suitable grasp configuration for systematically controlling the failure behavior of an object to break it at the desired location while avoiding damage on other areas. Our approach is based on failure task information and gripper wrench insertion capability. Failure task definition is accomplished using failure theories. Gripper wrench insertion capability is formulated by modeling the friction between the object and gripper. A new method inspired by human pre-manipulation process is introduced to utilize gripper itself as a friction measurement device. The provided friction model is capable of handling the anisotropic behavior of materials which is the case for fruits and vegetables. The evaluation method is formulated as a quasistatic grasp problem. Additionally, the general case of both fully-actuated and under-actuated grippers are considered. As a validation of the proposed evaluation method, experimental results for failing parts using Kuka Light-Weight Robot IV robot are presented.