ICRA 2022 论文列表

This paper proposes a parameterization method to represent SO (3) over multiple turns. This method is called a memory-based parameterization, because the idea is to integrate the past trajectory of exponential coordinates. The parameterization is consistent in the sense that the true rotation matrix can be reconstructed by using the exponential map. As an application of the proposed method, a 6D impedance controller is designed with a radially unbounded potential function. Consequently, in contrast to the conventional methods, an arbitrarily large angular deflection can be accommodated, resulting in a more realistic impedance behavior. The proposed schemes are validated through simulations and experiments.

This paper mathematically conceives a novel walking support device that leverages passive dynamics and coupling effects. In this model, a passive human walker is flexibly connected to an active humanoid, where the coupling effect induces a stable walking gait of the human. To understand the key mechanism of such indirect gait regulation, different actuation modes are designed for the humanoid and compared via phase-plane analysis of the steady-state gaits. Moreover, stability analysis is conducted via Poincaré map. The results show that it is difficult to enhance the human walker's stability when coupled to a humanoid robot using additional sensory information, compared to using a humanoid robot actuated with a predetermined force that employs no state feedback. The present mathematical model and our theoretical findings contribute to analysis and control design for locomotion systems with robot-human or inter-robots cooperation.

Damping properties in biological muscle are crit-ical for absorbing shock, maintaining posture, and positioning limbs and appendages. When creating biomimetic robots, the ability to replicate the dynamics of biological muscle is neces-sary to reproduce behaviors seen in an animal model. However, the damping properties of existing soft artificial muscles are difficult to predict and tune to match specific muscles as may be needed in biomimetic robots. Here, we present the design, manufacturing, and characterization of a novel soft damper to enable a greater degree of biomimetism in these soft actuators. The damper is composed of magnetorheological fluid contained within an elastomeric shell, which is cast using low-cost 3D printed parts and commercially available urethane rubber. We demonstrate that the force-velocity response over a velocity range of 0.1 to 10 mm/s is proportional to applied magnetic flux densities between 0.12 and 0.31 T. In the presence of a 0.31 T magnetic field from a small permanent magnet, the damper is capable of a maximum damping force increase of 13.2 N to 15.5 N relative to the 0 T control, at a compression depth of 7.9 mm, which is larger than that of several previously reported centimeter-scale dampers. As a proof-of-concept for integration with a Pneumatic Artificial Muscle (PAM), we use two parallel dampers to reduce the oscillations of a rapidly pressurized McKibben actuator. The ability to modulate the force-velocity performance of our elastomeric damper paves the way for custom damping profiles that can be used to improve biomimetism in soft robotic actuators.

Hip exoskeletons may hold potential to augment walking performance and mobility in individuals with disabilities. The purpose of this study was to design and validate a novel autonomous hip exoskeleton with a user-adaptive control strategy capable of reducing the energy cost of level and incline walking in individuals with and without walking impairment. First, in a small cohort of three unimpaired individuals, we validated the ability of our control strategy to provide hip flexion-extension torque that was proportional to the biological hip moment and reduce the energy cost of level and incline walking (24 ± 5% and 13 ± 5% reductions, respectively). Next, in a clinical feasibility experiment with an individual with significant walking impairment from cerebral palsy, we demonstrated that our untethered device and adaptive control scheme improved hip extension by 14° across the gait cycle, reduced average rectus femoris and semitendinosus muscle activity by 23% and 46%, respectively, and resulted in a 15% improvement in metabolic cost relative to walking without wearing the device.

Soft robotics is an emerging field that yields promising results for tasks that require safe and robust interactions with the environment or with humans, such as grasping, manipulation, and human-robot interaction. Soft robots rely on intrinsically compliant components and are difficult to equip with traditional, rigid sensors which would interfere with their compliance. We propose a highly flexible tactile sensor that is low-cost and easy to manufacture while measuring contact pressures independently from 14 taxels. The sensor is built from piezoresistive fabric for highly sensitive, continuous responses and from a custom-designed flexible printed circuit board which provides a high taxel density. From these taxels, location and intensity of contact with the sensor can be inferred. In this paper, we explain the design and manufacturing of the proposed sensor, characterize its input-output relation, evaluate its effects on compliance when equipped to the silicone-based pneumatic actuators of the soft robotic RBO Hand 2, and demonstrate that the sensor provides rich and useful feedback for learning-based in-hand object recognition.

Centroidal dynamics, which describes the overall linear and angular motion of a robot, is often used in locomotion generation and control of legged robots. However, the equation of centroidal dynamics contains nonlinear terms mainly caused by the robot's angular motion and needs to be linearized for deriving a linear model-predictive motion controller. This paper proposes a new linearization of the robot's centroidal dynamics. By expressing the angular motion with exponential coordinates, more linear terms are identified and retained than in the existing methods to reduce the loss from the model linearization. As a consequence, a model-predictive control (MPC) algorithm is derived and shows a good performance in tracking angular motions on a quadruped robot.

Event cameras are bio-inspired sensors that capture per-pixel asynchronous intensity change rather than the synchronous absolute intensity frames captured by a classical camera sensor. Such cameras are ideal for robotics applications since they have high temporal resolution, high dynamic range and low latency. However, due to their high temporal resolution, event cameras are particularly sensitive to flicker such as from fluorescent or LED lights. During every cycle from bright to dark, pixels that image a flickering light source generate many events that provide little or no useful information for a robot, swamping the useful data in the scene. In this paper, we propose a novel linear filter to preprocess event data to remove unwanted flicker events from an event stream. The proposed algorithm achieves over 4.6 times relative improvement in the signal-to-noise ratio when compared to the raw event stream due to the effective removal of flicker from fluorescent lighting. Thus, it is ideally suited to robotics applications that operate in indoor settings or scenes illuminated by flickering light sources. Code, Datasets and Video: https://github.com/ziweiWWANG/EFR

To facilitate sensing and physical interaction in remote and/or constrained environments, high-extension, lightweight robot manipulators are easier to transport and reach substantially further than traditional serial chain manipulators. We propose a novel planar 3-degree-of-freedom manipulator that achieves low weight and high extension through the use of a pair of spooling bistable tapes, commonly used in self-retracting tape measures, which are pinched together to form a reconfigurable revolute joint. The pinching action flattens the tapes to produce a localized bending region, resulting in a revolute joint that can change its orientation by cable tension and its location on the tapes though friction-driven movement of the pinching mechanism. We present the design, implementation, kinematic modeling, stiffness behavior of the revolute joint, and quasi-static performance of this manipulator. In particular, we demonstrate the ability of the manipulator to reach specified targets in free space, reach a 2D target with various orientations, and maintain an end-effector angle or stationary bending point while changing the other. The long-term goal of this work is to integrate the manipulator with an aerial robot to enable more capable aerial manipulation.

Grasping and manipulation are two of the most important hand functions that allow people to efficiently execute activities of daily living. Over the last years, many robotic devices have been proposed to assist people who suffer from neurological conditions by enhancing their grasping capabilities. In this work, we focus on the development of a robotic exoskeleton glove that can increase the grasp stability and the force exertion capabilities of the user by employing soft, telescopic, inflatable structures on the palmar side of the hand. Also, the proposed design employs a camera and an object identification system to facilitate the development of a shared control scheme that simplifies the operation of the device. The experiments demonstrate that the soft robotic exoskeleton glove can successfully execute semi-autonomous grasps and that the soft telescopic structures can increase the total exerted grasping forces by more than 40% when inflated.

Many possible fields of application of robots in real world settings hinge on the ability of robots to grasp objects. As a result, robot grasping has been an active field of research for many years. With our publication we contribute to the endeavor of enabling robots to grasp, with a particular focus on bin picking applications. Bin picking is especially challenging due to the often cluttered and unstructured arrangement of objects and the often limited graspability of objects by simple top down grasps. To tackle these challenges, we propose a fully self-supervised reinforcement learning approach based on a hybrid discrete-continuous adaptation of soft actor-critic (SAC). We employ parametrized motion primitives for pushing and grasping movements in order to enable a flexibly adaptable behavior to the difficult setups we consider. Furthermore, we use data augmentation to increase sample efficiency. We demonstrate our proposed method on challenging picking scenarios in which planar grasp learning or action discretization methods would face a lot of difficulties.

We describe a task and motion planning architecture for highly dynamic systems that combines a domain-independent sampling-based deliberative planning algorithm with a global reactive planner. We leverage the recent development of a reactive, vector field planner that provides guarantees of reachability to large regions of the environment even in the face of unknown or unforeseen obstacles. The reachability guarantees can be formalized using contracts that allow a deliberative planner to reason purely in terms of those contracts and synthesize a plan by choosing a sequence of reactive behaviors and their target configurations, without evaluating specific motion plans between targets. This reduces both the search depth at which plans will be found, and the number of samples required to ensure a plan exists, while crucially preserving correctness guarantees. The result is reduced computational cost of synthesizing plans, and increased robustness of generated plans to actuator noise, model misspecification, or unknown obstacles. Simulation studies show that our hierarchical planning and execution architecture can solve complex navigation and rearrangement tasks, even when faced with narrow passageways or incomplete world information.

Maneuvering an autonomous vehicle under drift condition is critical to the safety of autonomous vehicles when there is a sudden loss of traction due to external conditions such as rain or snow, which is a challenging control problem due to the presence of significant sideslip and nearly full saturation of the tires. In this paper, we focus on the control of drift maneuvers of autonomous vehicle to track circular paths with either fixed or moving centers, subject to change in the tire-ground interaction. In order to achieve the above tasks, we propose a hierarchical control architecture which decouples the curvature and center control of the trajectory. In particular, an outer control loop is proposed to stabilize the center by tuning the target curvature, and an inner control loop tracks the curvature using a feedforward/feedback controller enhanced by an \mathcal{L}_{1}\mathcal{L}_{1} adaptive component. The hierarchical architecture is flexible because the inner loop is task-agnostic and adaptive to changes in tire-ground interaction, which allows the outer loop to be designed independent of low-level dynamics, opening up the possibility of incorporating sophisticated planning algorithms. We implement our control strategy on a simulation platform as well as on a 1/10 scale RC car, and both the simulation and experiment results illustrate the effectiveness of our strategy in achieving the above described set of drift maneuvering tasks.

Graph Neural Networks (GNNs) are a paradigm-shifting neural architecture to facilitate the learning of complex multi-agent behaviors. Recent work has demonstrated remarkable performance in tasks such as flocking, multi-agent path planning and cooperative coverage. However, the policies derived through GNN-based learning schemes have not yet been deployed to the real-world on physical multi-robot systems. In this work, we present the design of a system that allows for fully decentralized execution of GNN-based policies. We create a framework based on ROS2 and elaborate its details in this paper. We demonstrate our framework on a case-study that requires tight coordination between robots, and present first-of-a-kind results that show successful real-world deployment of GNN-based policies on a decentralized multi-robot system relying on Adhoc communication. A video demonstration of this case-study can be found online11youtube.com/watch?v=COh-WLn4i04.

Grasping fragile objects in the presence of un-certainty is a crucial task for robots, that becomes inherently challenging if the manipulator in use is an industrial robot platform that does not provide compliant control inputs. This requires not only to estimate the alignment error during object contact but also to alter the robot configuration to decrease this error while taking interaction constraints into account. Thus, this work proposes a novel grasping controller tailored to industrial robots by exploiting tactile sensor feedback on the robot gripper fingers in order to estimate and compensate for the alignment error when touching the object. Specifically, we propose two grasping strategies, that allow to either directly compensate for interaction wrenches or to solve a model predictive control-problem to minimize the estimated alignment error. Eventually, we outline how these modalities can be realized as a hybrid Cartesian force-velocity-controller on an industrial manipulator. We evaluate the proposed grasping strategies on a WSG 50 parallel two-finger gripper, that is equipped with a digital sensor array (DSA) per finger, for which we also provide an extended ROS-driver that allows to obtain DSA-data at a communication rate above 5 Hz. Given the collected empirical evidence, the presented grasping controller increases the skill-set of industrial robots in the presence of uncertainty and thus allows to apply stiff robots to handle fragile objects autonomously.

Real-time and precise morphological analysis of intraoperative AAA is a significant pre-imperative for robot-assisted minimally invasive surgery (RMIS). However, this task is frequently accompanied by the difficulties of ambiguous boundaries and obscured surfaces of aneurysms. To remedy these problems, we propose a Light-Weight Dual-Stream Boundary-Aware Network (DSB-Net) and a novel diagnosis algorithm for real-time morphological analysis of AAA. In the network, the features at the boundaries are preserved by incorporating a boundary localization stream, while the interior segmentation accuracy is guaranteed with a mask prediction stream. Moreover, the diagnosis algorithm is developed to measure the exact size of AAA. Quantitative and qualitative assessments on two different types of datasets illustrate that (1) The presented DSB-Net remarkably outperforms the other previously proposed medical networks with the inference rate of 10.8 FPS, which meets the real-time clinical necessities. (2) The developed algorithm provides accurate size measurements for AAA, which indicates the proposed approach can be integrated into the robotic navigation framework for RMIS.

Next best view (NBV) is a technology that finds the best view sequence for sensor to perform scanning based on partial information, which is the core part for robot active reconstruction. Traditional works are mostly based on the evaluation of candidate views through time-consuming volu-metric transformation and ray casting, which heavily limits the applications of NBV. Recent deep learning based NBV methods aim to approximately learn the evaluation function by large-scale training, and improve both the effectiveness and efficiency of NBV. However, these methods force the network to regress the exact groundtruth value of each candidate view, which is much harder than simply ranking all the candidate views. Besides, most previous NBV works assume perfect sensing and perform in simulation environments, lacking real application abilities. In this paper, we propose a novel double branch NBV network, DB-NBV, to utilize the ranking process together with the evaluation process. We further design a real NBV robot and a pipeline to conduct real active reconstruction. Experiments on both simulation and real robot show that our method achieves the best performance and can be applied to real application with high accuracy and speed.

Clustering objects from the LiDAR point cloud is an important research problem with many applications such as autonomous driving. To meet the real-time requirement, existing research proposed to apply the connected-component-labeling (CCL) technique on LiDAR spherical range image with a heuristic condition to check if two neighbor points are connected. However, LiDAR range image is different from a binary image which has a deterministic condition to tell if two pixels belong to the same component. The heuristic condition used on the LiDAR range image only works empirically, which suggests the LiDAR clustering algorithm should be robust to potential failures of the empirical heuristic condition. To overcome this challenge, this paper proposes a divide-and-merge LiDAR clustering algorithm. This algorithm firstly conducts clustering in each evenly divided local region, then merges the local clustered small components by voting on edge point pairs. Assuming there are NN LiDAR points of objects in total with mm divided local regions, the time complexity of the proposed algorithm is O(N)+O(m^{2})O(N)+O(m^{2}). A smaller mm means the voting will involve more neighbor points, but the time complexity will become larger. So the mm controls the trade-off between the time complexity and the clustering accuracy. A proper mm helps the proposed algorithm work in real-time as well as maintain good performance. We evaluate the divide-and-merge clustering algorithm on the SemanticKITTI panoptic segmentation benchmark by cascading it with a state-of-the-art semantic segmentation model. The final performance evaluated through the leaderboard achieves the best among all published methods. The proposed algorithm is implemented with C++ and wrapped as a python function. It can be easily used with the modern deep learning framework in python. We released the code under the following link 11https://github.com/placeforyiming/Divide-and-Merge-LiDAR-Panoptic-Cluster. Show More

In this paper, we propose a detumbling strategy that stabilizes the motion of a tumbling client satellite using an orbital servicing manipulator, which is the goal of the post-grasp phase. One of the critical aspects in this phase is ensuring that excessive contact forces are not generated at the grasp interface. In addition, space mission requirements might demand a nominal manipulator configuration that is suitable for further manipulation/servicing activities. The proposed strategy allows the detumbling of the client motion while ensuring that the contact forces developed at the grasp interface do not violate a safety threshold. Further, it allows the reconfiguration of the manipulator arm by exploiting the full actuation capability of the manipulator-equipped servicing spacecraft. The controller guarantees joint task convergence in the nullspace of the manipulator's end-effector, and is also valid for kinematically singular configurations of the manipulator. It is further augmented using a quadratic programming based approach to optimally constrain the contact forces. Finally, simulation results for a post-grasp detumbling scenario are shown to validate the effectiveness of the proposed method.

Particle robots are novel biologically-inspired robotic systems where locomotion can be achieved collectively and robustly, but not independently. While its control is currently limited to a hand-crafted policy for basic locomotion tasks, such a multi-robot system could be potentially controlled via Deep Reinforcement Learning (DRL) for different tasks more efficiently. However, the particle robot system presents a new set of challenges for DRL differing from existing swarm robotics systems: the low degrees of freedom of each robot and the increased necessity of coordination between robots. We present a 2D particle robot simulator using the OpenAI Gym interface and Pymunk as the physics engine, and introduce new tasks and challenges to research the underexplored applications of DRL in the particle robot system. Moreover, we use Stable-baselines3 to provide a set of benchmarks for the tasks. Current baseline DRL algorithms show signs of achieving the tasks but are yet unable to reach the performance of the hand-crafted policy. Further development of DRL algorithms is necessary in order to accomplish the proposed tasks.