ICRA 2021 Papers

MRI-guided robotic systems are emerging platforms for minimally invasive intervention because of high positioning accuracy and excellent tissue contrast. MR safe encoders are critical components for closed-loop robotic control. This paper develops an MR safe absolute rotary encoder based on eccentric sheave and FBG sensors. The eccentric sheave transforms the rotational motion of the shaft to the bending deflection of the beam on which FBG sensors are integrated. A model is built by establishing the relationship of the kinematics of the sheave, the mechanical properties of the beam with unknown length, and the strain model of two Fiber Bragg Grating (FBG) sensors. A Pseudo-Rigid Body (PRB) 3R model is used to solve a set of constrained equations for accurate rotary encoding. A prototype is built to calibrate the parameters and validate the accuracy of the encoder and its MR compatibility. Results show that the maximum angular error is 1.6°, and the RMS error is 0.46°. MRI shows that no noticeable artifacts are observed, and the Signal to Noise Ratio (SNR) is not affected. The results demonstrate the potential of the proposed method for it to be integrated with MR safe robots with easy fabrication, compact structures, and continuous measurement.

In human-in-the-loop navigation, the operator’s intention is to locally avoid obstacles while planning long-horizon paths in order to complete the navigation task. We propose a hierarchical teleoperation framework that captures these characteristics of intention, and generates trajectories that are locally safe and follow the operator’s global plan. The hierarchical teleoperation framework consists of 1) a global path which encapsulates the intended direction of the operator, 2) local trajectories that circumvent obstacles near the vehicle’s vicinity while following the global path, and 3) safety monitoring to avoid possible imminent collisions. By removing the operator from providing dynamic-level control inputs and instead having inputs inform trajectory generation, we show a significant reduction of the operator’s engagement while maintaining smooth performance.We showcase hierarchical teleoperation in navigation tasks in a random forest environment and a high-clutter warehouse characterized by narrow gaps and dense obstacles. With our method, we maintain consistent high speed throughout the task with smooth jerk profiles, decreased time to completion, and significantly reduced operator engagement.

The barcode navigation based on QR (quick response) codes is widely employed in industrial logistics due to its accurate localization and flexible movement paths. However, the regular repair of damaged barcodes and robot speed control when approaching the barcodes are required. In this study, we presented an improved magnetic spot navigation approach to replace the barcode navigation for automated guided vehicles (AGVs). The fusion of the high-precision magnetic tracking method and odometer based on AGV encoders can overcome the disadvantages of barcode navigation. The magnetic tracking approach provides the AGV pose relative to the nearest magnet spot, instead of the low-precision longitudinal and lateral measurement via a magnetic ruler. Besides, with the benefit of the adaptive weighted fusion algorithm, the distance between the adjacent barcode can be set from 500 to 1000 mm via magnetic spots. Experimental results show that the mean path accuracy and mean magnet spot localization accuracy of the improved magnetic spot navigation were 110 ± 30 mm and 14.5 ± 0.87 mm, respectively. The proposed approach provides a novel possibility for large-area and high-precision navigation in AGVs-based industrial logistics, especially for large outdoor scenarios.

a new generation of computer vision, namely event-based or neuromorphic vision, provides a new paradigm for capturing visual data and the way such data is processed. Due to a highly novel type of visual sensors used in event-based vision, only a few datasets aimed at visual navigation tasks are publicly available.In this paper, we present and describe the first event-based vision dataset intended to cover visual navigation tasks for mobile robots navigating in different types of agricultural environment. The dataset might open new opportunities for the evaluation of existing and creation of new event-based visual navigation methods for use in agricultural scenes that contain a lot of vegetation, animals, and patterned objects.The new dataset was created using our own custom-designed Sensor Bundle, which was installed on a mobile robot platform. During data acquisition sessions, the platform was manually controlled. The Sensor Bundle consists of a dynamic vision sensor, a LIDAR, an RGB-D camera, and environmental sensors.In total, 21 data sequences in 12 different scenarios for the autumn season are publicly available. Each data sequence is accompanied by a video demonstrating its content and a detailed description, including known issues. The new dataset is mostly designed for Visual Odometry tasks; however, it also includes loop-closures for applying event-based visual SLAM methods.

Visual Inertial Odometry (VIO) is of great interest due the ubiquity of devices equipped with both a monocular camera and Inertial Measurement Unit (IMU). Methods based on the extended Kalman Filter remain popular in VIO due to their low memory requirements, CPU usage, and processing time when compared to optimisation-based methods. In this paper, we analyse the VIO problem from a geometric perspective and propose a novel formulation on a smooth quotient manifold where the equivalence relationship is the well-known invariance of VIO to choice of reference frame. We propose a novel Lie group that acts transitively on this manifold and is compatible with the visual measurements. This structure allows for the application of Equivariant Filter (EqF) design leading to a novel filter for the VIO problem. Combined with a very simple vision processing front-end, the proposed filter demonstrates state-of-the-art performance on the EuRoC dataset compared to other EKF-based VIO algorithms.

This paper focuses on developing a real-time and flexible velocity estimation approach for serial revolute manipulator using only one inertial measurement unit (IMU) mounted on each link side of the manipulator. Particularly, the proposed approach has no requirement for the installation position and orientation of the IMU, which improves the flexibility of the implementation procedure. A joint velocity model is established based on the proposed principle of constructing coordinate system according to the robotic geometric information. The general solutions are derived in detail, thereby the proposed algorithm can be generalized into any other robots with the same geometric configuration. With the method, the joint rotational velocity measurements of static and dynamic robotic motion are provided compared to encoders. Experimental results based on the six degrees of freedom (DOF) collaborative manipulator have validated the feasibility and effectiveness of the proposed approach. The proposed method has the benefits of low cost and flexibility, which could work as a redundant velocity monitor criterion to provide assistant joint velocity measurements.

Self-assembly has attracted growing interests in modular robotics during past decades. Recent work accelerates the assembly process by parallelizing the docking actions among robots. However, these methods can only apply to ideal environments without obstacles. Otherwise, robots will get trapped during the assembly process, due to the complex scenes with obstacles. This paper presents an efficient parallel assembly planning algorithm for modular robots by taking the surrounding obstacles into consideration. By this algorithm, the docking actions are able to avoid immovable obstacles, and therefore parallel self-assembly of robots can adapt to complex environments. To validate the efficacy and generality, the authors have implemented this algorithm in a grid-world simulation environment with 25 distinct maps. The simulation results show a much higher success rate (more than 80%) of our proposed algorithm compared with the existing parallel self-assembly planning algorithms. Finally, the feasibility of the algorithm is affirmed by a self-assembly experiment on the automated guided vehicles (AGVs).

This paper derives a closed-form method for computing hybrid force-velocity control. The key idea is to maximize the kinematic conditioning of the mechanical system, which includes a robot, free objects, a rigid environment and contact constraints. The method is complete, in that it always produces an optimal/near optimal solution when a solution exists. It is efficient, since it is in closed form, avoiding the iterative search of our previous work. We test the method on 78,000 randomly generated test cases. The method outperforms our previous search-based technique by being from 7 to 40 times faster, while consistently producing better solutions in the sense of robustness to kinematic singularity. We also test the method in several representative manipulation experiments.

Research in autonomous robots for construction has largely focused on ground-based robots whose reach constrains the size of what they can build, or on climbing or aerial robots that build solid or unroofed structures. Autonomous construction of larger, multistory buildings, or bridges spanning unsupported distances, would require robots that build sturdy structures supporting their own weight. In this paper, we present VaultBot, a system of autonomous robots that build a load-bearing spanning vault using identical modular blocks. The custom blocks employ mechanical and other features to facilitate robotic manipulation and locomotion, and can be removed from and replaced in an assembled structure as a way of repairing damage. We characterize the system's performance and failure modes, and demonstrate reliable autonomous assembly for a structure composed of 46 blocks. Blocks can be made collapsible and deployable as a way of reducing mass and volume that must be transported to a construction site. Such a system could be used to help enable construction of protective shelters in challenging environments, such as disaster relief scenarios, arctic settings, or extraterrestrial habitats.

In this paper, we proposed an autonomous robotic flexible endoscope system for the laparoscopic bariatric surgery (LBS). This system comprises a UR5 robot and a flexible endoscope equipped with a novel continuum joint, named reinforced double helix continuum mechanism. Compared with the simple helix structure, the compressional and torsional stiffness of the proposed joint are improved significantly. To automate the robotic flexible endoscope, image-based visual servoing technique is employed. A deep learning algorithm named TernausNet-16 is improved and incorporated into the control framework to detect surgical instruments inside the camera view. The experimental studies verified the effectiveness and feasibility of the robotic flexible endoscope system for the visual serviong control scheme assisted by deep learning methods.

Industrial knitting machines create fabric by manipulating loops held on hundreds of needles. A core problem in pattern making for these machines is transfer planning – coming up with a sequence of low-level operations that move loops to the appropriate needles so that knitting through those loops produces the correct final structure. Since each loop is connected to the larger piece in progress, transfer plans must account for not only loop position, but the way strands of yarn tangle around each other.We present the first complete, discrete representation of the machine’s loop-tangling process. Our representation combines a braid from the Artin Braid Group with an array of explicit loop positions to fully capture loop crossings. By storing braids in the Symmetric Normal Form, states can be quickly compared and updated incrementally with machine operations. This representation can be used to verify the equivalence of transfer operations, providing an important tool in optimizing knit manufacturing.We improve on prior work in transfer planning algorithms, which can only solve certain subclasses of problems and are frequently suboptimal in terms of fabrication time, by introducing a novel A* search heuristic and state-collapsing mechanism, which we show finds optimal transfer plans for a large benchmark set of small transfer planning problems.

In this article, we introduce a cooperative path planning algorithm for a cardinal and a support robot where the cardinal robot is unable to traverse a subset of edges in a network until the support robot has first traversed them. This subset of edges represent paths in an environment that are initially unavailable to the cardinal robot and require the assistance of the support robot. A (2 + α)-approximation algorithm (where α is the supremum of the ratio of the travel time of the support robot versus the travel time of the cardinal robot) is presented for this problem and is applied to various types of networks in order to examine the quality of the solutions it produces. We then conclude by discussing some potential future work concerning variations of this problem.

Aircraft routing problem is a crucial component for flight automation. Despite recent successes, challenges still remain when the environment is dynamic and uncertain. In this paper, we tackle the following two challenges. First, when the environment is uncertain, it is much safer if the route planner can guarantee a specified level of safety. Second, when the environment is dynamic, the planner needs to adapt to the changes in the environment quickly. To address these challenges, we present three contributions. First, we propose formulating the aircraft routing problem under a dynamic and uncertain environment as a chance constrained stochastic shortest path (CC-SSP) problem. Second, we introduce an anytime algorithm for the CC-SSP problem, which is effective in a dynamic environment with limited planning time. To be more specific, we present two versions of the algorithm and compare their performances. Third, we show that the algorithm can be generalized to solve a larger class of problems called chance constrained partially observable Markov decision process (CC-POMDP).

The flock-guidance problem enjoys a challenging structure where multiple optimization objectives are solved simultaneously. This usually necessitates different control approaches to tackle various objectives, such as guidance, collision avoidance, and cohesion. The guidance schemes, in particular, have long suffered from complex tracking-error dynamics. Furthermore, techniques that are based on linear feedback strategies obtained at equilibrium conditions either may not hold or degrade when applied to uncertain dynamic environments. Pre-tuned fuzzy inference architectures lack robustness under such unmodeled conditions. This work introduces an adaptive distributed technique for the autonomous control of flock systems. Its relatively flexible structure is based on online fuzzy reinforcement learning schemes which simultaneously target a number of objectives; namely, following a leader, avoiding collision, and reaching a flock velocity consensus. In addition to its resilience in the face of dynamic disturbances, the algorithm does not require more than the agent position as a feedback signal. The effectiveness of the proposed method is validated with two simulation scenarios and benchmarked against a similar technique from the literature.

In-hand manipulation is challenging for soft robotic hands, especially in the real world where robots encounter a variety of object sizes and shapes. As such, the role of the palm is crucial, providing stabilizing contact to objects during grasping and manipulation, and controlling the position of objects with respect to the fingertips. We demonstrate an actuated palm capable of enhancing the in-hand manipulation capabilities of a soft hand by better-utilizing limited finger dexterity. With a combination of physical and virtual experiments, we explore the effects of palm diameter and height on in-hand manipulation performance over a variety of object shapes and sizes, and three key manipulation primitive motions. The results of these experiments show that maintaining manipulation capabilities over a large range of object sizes requires the palm’s diameter to decrease as a function of its height to prevent interference between the fingers and palm. Based on these insights, we design an actuated palm mechanism that achieves the desired relationship between palm height and diameter using one actuated degree of freedom. Finally, we show that this adjustable palm enables the hand to manipulate a larger range of object sizes and aspect ratios, and its utility is demonstrated in a mid-air shelving in-hand manipulation task.

Variable stiffness technology is an extensively discussed topic in soft robotics, which can bridge the traditional fast, precise and high-force rigid robots with compliant, agile, and safe soft robots. In this paper, we introduce the concept of geometry-induced rigidity and propose a variable curvature jamming (VCJ) mechanism for amplifying the laminar jamming structures which are fast, efficient, low cost, easy control, and very versatile in stiffness modulation applications. By adjusting the cross sectional curvature of the laminar jamming structures, and coupling with the vacuum jamming, the initial compliant laminar structures can stiffen notably. The bidirectional stiffness behaviors of the VCJ actuators with different design parameters are experimentally studied and analyzed. The results show that the transverse curvature can remarkably enhance the stiffness and load capacity of the actuators, about 3 folds in stiffness change than the laminar jamming mechanism. The VCJ actuator can even bear 10.85-kg weight which is about 1700 times of its own weight at its rigid state. And it can be operated in three stiffening modes with different degrees of variable stiffness performance. Moreover, the proposed VCJ mechanism is compatible with all existing laminar jamming designs, which provide an effective design method for building soft actuators and soft robots with excellent variable stiffness performance and load capacity.

Well-established optimization-based methods can guarantee an optimal trajectory for a short optimization horizon, typically no longer than a few seconds. As a result, choosing the optimal trajectory for this short horizon may still result in a sub-optimal long-term solution. At the same time, the resulting short-term trajectories allow for effective, comfortable and provable safe maneuvers in a dynamic traffic environment. In this work, we address the question of how to ensure an optimal long-term driving strategy, while keeping the benefits of classical trajectory planning. We introduce a Reinforcement Learning based approach that coupled with a trajectory planner, learns an optimal long-term decision-making strategy for driving on highways. By online generating locally optimal maneuvers as actions, we balance between the infinite low-level continuous action space, and the limited flexibility of a fixed number of predefined standard lane-change actions. We evaluated our method on realistic scenarios in the open-source traffic simulator SUMO and were able to achieve better performance than the 4 benchmark approaches we compared against, including a random action selecting agent, greedy agent, high-level, discrete actions agent and an IDM-based SUMO-controlled agent.

In many applications, including logistics and manufacturing, robot manipulators operate in semi-structured environments alongside humans or other robots. These environments are largely static, but they may contain some movable obstacles that the robot must avoid. Manipulation tasks in these applications are often highly repetitive, but require fast and reliable motion planning capabilities, often under strict time constraints. Existing preprocessing-based approaches are beneficial when the environments are highly-structured, but their performance degrades in the presence of movable obstacles, since these are not modelled a priori. We propose a novel preprocessing-based method called Alternative Paths Planner (APP) that provides provably fixed-time planning guarantees in semi-structured environments. APP plans a set of alternative paths offline such that, for any configuration of the movable obstacles, at least one of the paths from this set is collision-free. During online execution, a collision-free path can be looked up efficiently within a few microseconds. We evaluate APP on a 7 DoF robot arm in semi-structured domains of varying complexity and demonstrate that APP is several orders of magnitude faster than state-of-the-art motion planners for each domain. We further validate this approach with real-time experiments on a robotic manipulator.