ICRA 2020 Papers

This study describes the enhancement of a vacuum actuated soft gripper's grasping capabilities using retractable finger nails and an active re-configurable palm. The finger nail mechanism is pneumatically actuated and enables the gripper to perform complex grasping and manipulation tasks with high repeatability. The retracted nails can exert normal grasping forces of up to 1.8N and enable grasping of objects up to 200μm thick from flat surfaces, while allowing the gripper to execute delicate pinch grasps without complex trajectory or grasp planning. A wide array of robotic grasping tasks that were not possible without nails are also described.

Applications of deep neural network (DNN) based object and grasp detections could be expanded significantly when the network output is processed by a high-level reasoning over relationship of objects. Recently, robotic grasp detection and object detection with reasoning have been investigated using DNNs. There have been efforts to combine these multitasks using separate networks so that robots can deal with situations of grasping specific target objects in the cluttered, stacked, complex piles of novel objects from a single RGB-D camera. We propose a single multi-task DNN that yields accurate detections of objects, grasp position and relationship reasoning among objects. Our proposed methods yield state-of-the-art performance with the accuracy of 98.6% and 74.2% with the computation speed of 33 and 62 frame per second on VMRD and Cornell datasets, respectively. Our methods also yielded 95.3% grasp success rate for novel object grasping tasks with a 4-axis robot arm and 86.7% grasp success rate in cluttered novel objects with a humanoid robot.

Over the past few years, shape memory polymer (SMP) has been extensively studied in terms of its remarkable reversible dry adhesive properties and related smart adhesive applications. However, its exceptional properties have not been exploited for further opportunities such as pick-and-place applications, which would otherwise advance the robotic manipulation. This work explores the use of an SMP to design an adhesive gripper that picks and places a target solid object employing the reversible dry adhesion of an SMP. Compared with other single surface contact grippers including vacuum, electromagnetic, electroadhesion, and gecko grippers, the SMP adhesive gripper interacts with not only flat and smooth dry surfaces but also moderately rough and even wet surfaces for pick-and-place with high adhesion strength (> 2 atmospheres). In this work, associated physical mechanisms, SMP adhesive mechanics, and thermal conditions are studied. In particular, the numerical and experimental study elucidates that the optimal compositional and topological SMP design may substantially enhance its adhesion strength and reversibility, which leads to a strong grip force simultaneously with a minimized releasing force. Finally, the versatility and utility of the SMP adhesive gripper are highlighted through diverse pick-and-place demonstrations.

Executing multiple tasks concurrently is important in many robotic applications. Moreover, the prioritization of tasks is essential in applications where safety-critical tasks need to precede application-related objectives, in order to protect both the robot from its surroundings and vice versa. Furthermore, the possibility of switching the priority of tasks during their execution gives the robotic system the flexibility of changing its objectives over time. In this paper, we present an optimization-based task execution and prioritization framework that lends itself to the case of time-varying priorities as well as variable number of tasks. We introduce the concept of extended set-based tasks, encode them using control barrier functions, and execute them by means of a constrained-optimization problem, which can be efficiently solved in an online fashion. Finally, we show the application of the proposed approach to the case of a redundant robotic manipulator.

This paper describes the design and implementation of a real-time robotics system for semi-specular/painted surface defect detection. The system can be used on moving parts, tolerate varying lighting conditions, and can accommodate small inherent vibrations of the inspected surface that is common in manufacturing operations. Topographical information of the inspected surface is first obtained by the analysis of reflections of a known pattern from this surface. Spectral analysis is then applied to identify defects through novelty detection. Finally, a defect tracking mechanism eliminates spurious defects. The proposed system operates continuously at 90 fps. The paper presents field testing results that show the system can be used as a consistent and cost-effective way of quality control.

In the context of model-based reinforcement learning and control, a large number of methods for learning system dynamics have been proposed in recent years. The purpose of these learned models is to synthesize new control policies. An important open question is how robust current dynamics-learning methods are to shifts in the data distribution due to changes in the control policy. We present a real-robot dataset which allows to systematically investigate this question. This dataset contains trajectories of a 3 degrees-of-freedom (DOF) robot being controlled by a diverse set of policies. For comparison, we also provide a simulated version of the dataset. Finally, we benchmark a few widely-used dynamics-learning methods using the proposed dataset. Our results show that the iid test error of a learned model is not necessarily a good indicator of its accuracy under control policies different from the one which generated the training data. This suggests that it may be important to evaluate dynamics-learning methods in terms of their transfer performance, rather than only their iid error.

This paper addresses the problem to simulate ARVA sensors using ROS and Gazebo. ARVA is a French acronym which stands for Appareil de Recherche de Victims en Avalanche and represents the forefront technology adopted in Search & Rescue operations to localize victims of avalanches buried under the snow. The aim of this paper is to describe the mathematical and theoretical background of the transceiver, discussing its implementation and integration with ROS allowing researchers to develop faster and smarter Search &Rescue strategies based on ARVA receiver data. To assess the effectiveness of the proposed sensor model, We present a simulation scenario in which an Unmanned Aerial Vehicle equipped with the transceiver sensor performs a basic S&R pattern using the output of ARVA system.

Logic-Geometric Programming (LGP) is a powerful motion and manipulation planning framework, which represents hierarchical structure using logic rules that describe discrete aspects of problems, e.g., touch, grasp, hit, or push, and solves the resulting smooth trajectory optimization. The expressive power of logic allows LGP for handling complex, large-scale sequential manipulation and tool-use planning problems. In this paper, we extend the LGP formulation to stochastic domains. Based on the control-inference duality, we interpret LGP in a stochastic domain as fitting a mixture of Gaussians to the posterior path distribution, where each logic pro le defines a single Gaussian path distribution. The proposed framework enables a robot to prioritize various interaction modes and to acquire interesting behaviors such as contact exploitation for uncertainty reduction, eventually providing a composite control scheme that is reactive to disturbance.

The advanced robotic and automation (ARA) lab has developed and successfully implemented a design inspired by many of the various cutting edge steel inspection robots to date. The combination of these robots concepts into a unified design came with its own set of challenges since the parameters for these features sometimes conflicted. An extensive amount of design and analysis work was performed by the ARA lab in order to find a carefully tuned balance between the implemented features on the ARA robot and general functionality. Having successfully managed to implement this conglomerate of features represents a breakthrough to the industry of steel inspection robots as the ARA lab robot is capable of traversing most complex geometries found on steel structures while still maintaining its ability to efficiently travel along these structures; a feat yet to be done until now.

Interactive simulators are used in several important applications which include the training simulators for teleoperated robotic laparoscopic surgery. While stateof-art simulators are capable of rendering realistic visuals and accurate dynamics, grasping is often implemented using kinematic simplification techniques that prevent truly multimanual manipulation, which is often an important requirement of the actual task. Realistic grasping and manipulation in simulation is a challenging problem due to the constraints imposed by the implementation of rigid-body dynamics and collision computation techniques in state-of-the-art physics libraries. We present a penalty based parametric approach to achieve multi-manual grasping and manipulation of complex objects at arbitrary postures in a real-time dynamic simulation. This approach is demonstrated by accomplishing multi-manual tasks modeled after realistic scenarios, which include the grasping and manipulation of a two-handed screwdriver task and the manipulation of a deformable thread.

We propose a serpentine type tendon driven underactuated end-effector design with a closing mechanism that is triggered upon contact with an object. This end-effector can grasp objects without knowing the size a priori and is able to grasp a new object while securing another one previously grasped, and so grasp multiple objects sequentially with a single DOF actuation. Design parameters based on the object dimensions are proposed. A low-cost prototype demonstrates two implementations (radius estimation and autonomous grasp of circular objects by torque control, and sequential grasps of multiple objects) of the end-effector through several experiments. A method for estimating applied internal forces is also proposed. This end-effector can benefit robotic manipulation in tasks such as fetching applications, industrial pick-and-place of single or multiple objects and human-robot hand-off interactions.

Traditional solar trackers often adopt motors to automatically adjust the attitude of the solar panels towards the sun for maximum power efficiency. In this paper, a novel design of solar tracker for the ocean environment is introduced. Utilizing the fluctuations due to the waves, electromagnetic brakes are utilized instead of motors to adjust the attitude of the solar panels. Compared with the traditional solar trackers, the proposed one is simpler in hardware while the harvesting efficiency is similar. The desired attitude is calculated out of the local location and time. Then based on the dynamic model of the system, the angular acceleration of the solar panels is estimated and a control algorithm is proposed to decide the release and lock states of the brakes. In such a manner, the adjustment of the attitude of the solar panels can be achieved by using two brakes only. Experiments are conducted to validate the acceleration estimator and the dynamic model. At last, the feasibility of the proposed solar tracker is tested on the real water surface. The results show that the system is able to adjust 40° in two dimensions within 28 seconds.

This paper presents a novel portable passive lower limb exoskeleton for walking assistance. The exoskeleton is designed with built-in spring mechanisms at the hip and knee joints to realize gravity balancing of the human leg. A pair of mating gears is used to convert the tension force from the built-in springs into balancing torques at hip and knee joints for overcoming the influence of gravity. Such a design makes the exoskeleton has a compact layout with small protrusion, which improves its safety and user acceptance. In this paper, the design principle of gravity balancing is described. Simulation results show a significant reduction of driving torques at the limb joints. A prototype of single leg exoskeleton has been constructed and preliminary test results show the effectiveness of the exoskeleton.

Orientability is an important performance index to evaluate the dexterity of haptic master devices. Currently, most of the existing haptic master devices have limited workspace and limited dexterity. In this paper, we present the RemoT-ARM, a 6 Degree-of-Freedom (DOF) haptic master device that can provide larger and more dexterous workspace for operators. To evaluate its reachability of orientations, we propose a novel orientability index. Furthermore, a relative orientability index is proposed to characterize the matching degree of the workspace of a given manipulator to its target workspace. The volume, the manipulability and the condition number are also introduced as performance indices to evaluate the size and the isotropy of the workspace. According to these performance indices, all possible configurations for the RemoT-ARM have been taken into consideration, analyzed, and compared to finalize its optimal configuration.

Fusions of LiDARs (light detection and ranging) and cameras have been effectively and widely employed in the communities of autonomous vehicles, virtual reality and mobile mapping systems (MMS) for different purposes, such as localization, high definition map or simultaneous location and mapping. However, the extrinsic calibration between a camera and a 3D LiDAR is a fundamental prerequisite to guarantee its performance. Some previous methods are inaccurate, have calibration error that is several times the beam divergence, and often require special calibration objects, thereby limiting their ubiquitous use for calibration. To overcome these shortcomings, we propose a novel and high-accuracy method for the extrinsic calibration between a camera and a 3D LiDAR. Our approach relies on the infrared images from a camera with an infrared filter, and the 2D-3D corresponding points in a scene with the corners of a wall can be extracted to calculate the six extrinsic parameters. Experiments using the Velodyne VLP-16 sensor show that the method can achieve an extrinsic accuracy at the level of the beam divergence, which is fully analyzed and validated from two different aspects. Therefore, the calibration method in this paper is highly accurate, effective and does not require special complicated calibration objects; thus, it meets the requirements of practical applications.

Soft robot has demonstrated promise in unstructured and dynamic environments due to unique advantages, such as safe interaction, adaptiveness, easy to actuate, and easy fabrication. However, the highly dissipative nature of elastic materials results in small stiffness of soft robot which limits certain functions, such as force transmission, position accuracy, and load capability. In this paper, we present a novel articulated soft robot with variable stiffness. The robot is constructed by rigid joints and compliant bistable structures in series. Each joint can be independently locked through triggering the bistable structure to touch the mechanical constrain. Thus, the bending stiffness of the joint can be magnified which increases the stiffness of the articulated soft robot. Through this construction method, even driven by only one servomotor, the robot demonstrates variable workspace and stiffness which have the potential of dexterous manipulation and maintaining shape under tip load.

This paper proposes a new path planning algorithm to consider motion uncertainty for wheeled robots in rough environments. The proposed method uses particles to express the uncertainty propagation in complicated environments constructed with various types of terrain. Also, RRT (Rapidly-exploring Random Tree) is expanded based on the uncertainty of each node in order to prevent increasing the accumulated position uncertainty. As a result, the generated path reduces the times of path-following and re-planning based on inaccurate localization information. The effectiveness of the proposed method is evaluated in simulation using motion uncertainty models obtained by experiments. The results show that the proposed method decreases the position uncertainty while keeping the probability to avoid collisions and to reach the goal area compared with conventional approaches.

The increasingly-used robotic systems can provide precise delivery and reduce X-ray radiation to medical staff in percutaneous coronary interventions (PCI), but natural manipulations of interventionalists are forgone in most robot-assisted procedures. Therefore, it is necessary to explore natural manipulations to design more advanced human-robot interfaces (HRI). In this study, a multilayer-multimodal fusion architecture is proposed to recognize six typical subpatterns of guidewire manipulations in conventional PCI. The synchronously acquired multimodal behaviors from ten subjects are used as the inputs of the fusion architecture. Six classification-based and two rule-based fusion algorithms are evaluated for performance comparisons. Experimental results indicate that the multimodal fusion brings significant accuracy improvement in comparison with single-modal schemes. Furthermore, the proposed architecture can achieve the overall accuracy of 96.90%, much higher than that of a singlelayer recognition architecture (92.56%). These results have indicated the potential of the proposed method for facilitating the development of HRI for robot-assisted PCI.