IROS 2020 论文列表

Despite the recent progress, guidance, navigation, and control (GNC) are largely unsolved for agile micro autonomous underwater vehicles (μAUVs). Hereby, robust and accurate self-localization systems which fit μAUVs play a key role and their absence constitutes a severe bottleneck in micro underwater robotics research. In this work we present, first, a small-size low-cost high performance vision-based self-localization module which solves this bottleneck even for the requirements of highly agile robot platforms. Second, we present its integration into a powerful GNC-framework which allows the deployment of μAUVs in fully autonomous mission. Finally, we critically evaluate the performance of the localization system and the GNC-framework in two experimental scenarios.

Soft materials with embedded magnetic properties can be actuated in a contactless manner for dexterous motion in restricted and unstructured environments. Magnetic soft robots have been demonstrated to be capable of versatile and programmable untethered motion. However, magnetic soft robots reported in literature are typically actuated by utilizing magnetic fields to generate torques that produce deformation. By contrast, this work investigates the utilization of field gradients to produce tethering forces for anchoring soft robots to the working surface, in conjunction with the use of magnetic fields to generate torques for deformation. The methodology applied here uses a six-coil electromagnetic system for field generation. The approach to achieve the magnetic field and gradients desired for soft robot motion is described, along with the restrictions imposed by Maxwell's equations. The design and fabrication of the soft robots is explained together with calculations to assess the capabilities of the actuation system. Proof-of-concept demonstrations of soft robot motion show Hexapede robots with the ability to `walk' untethered on the ceiling of the workspace, working against gravity; and lightweight Worm robots made of thin strips of material are demonstrated to locomote while staying in contact with the ground.

We present a general approach for constructing proofs of motion planning infeasibility. Effective high-dimensional motion planners, such as sampling-based methods, are limited to probabilistic completeness, so when no plan exists, these planners either do not terminate or can only run until a timeout. We address this completeness challenge by augmenting a sampling-based planner with a method to create an infeasibility proof in conjunction with building the search tree. An infeasibility proof is a closed polytope that separates the start and goal into disconnected components of the free configuration space. We identify possible facets of the polytope via a nonlinear optimization procedure using sampled points in the non-free configuration space. We identify the set of facets forming the separating polytope via a linear constraint satisfaction problem. This proof construction is valid for general (i.e., non-Cartesian) configuration spaces. We demonstrate this approach on the low-dimensional Jaco manipulator and discuss engineering approaches to scale to higher dimensional spaces.

A high-quality free-motion rendering is one of the most vital traits to achieve an immersive human-robot interaction. Rendering free-motion is notably challenging for rehabilitation exoskeletons due to their relatively high weight and powerful actuators required for strength training and support. In the presence of dynamic human movements, accurate feedback linearization of the robot's dynamics is necessary to allow for a linear synthesis of interaction wrench controllers. Hence, we introduce a virtual model controller that uses two 6-DoF force sensors to control the interaction wrenches of a multi-DoF torque-controlled exoskeleton over the joint accelerations and inverse dynamics. Furthermore, we propose a disturbance observer for controlling the joint acceleration to diminish the influence of modeling errors on the inverse dynamics. To provide a high-bandwidth, low-bias estimation of the system's acceleration, we introduce a bias-observer which fuses the information from joint encoders and seven low priced IMUs. We have validated the performance of our proposed control structure on the shoulder and arm exoskeleton ANYexo. The experimental comparison of the controllers shows a reduction of the felt inertia and maximum reflected joint torque by a factor of more than three compared to state of the art. The controllers' robustness w.r.t. a model mismatch is validated. The experiments show that the closed-loop acceleration control improves the tracking, particularly at joints with low inertia. The proposed controllers' performance sets a new benchmark in haptic transparency for comparable devices and should be transferable to other applications.

In this work, we present a pragmatic approach to enable unmanned aerial vehicle (UAVs) to autonomously perform highly complicated tasks of object pick and place. This paper is largely inspired by challenge-2 of MBZIRC 2020 and is primarily focused on the task of assembling large 3D structures in outdoors and GPS-denied environments. Primary contributions of this system are: (i) a novel computationally efficient deep learning based unified multi-task visual perception system for target localization, part segmentation, and tracking, (ii) a novel deep learning based grasp state estimation, (iii) a retracting electromagnetic gripper design, (iv) a remote computing approach which exploits state-of-the-art MIMO based high speed (5000Mb/s) wireless links to allow the UAVs to execute compute intensive tasks on remote high end compute servers, and (v) system integration in which several system components are weaved together in order to develop an optimized software stack. We use DJI Matrice-600 Pro, a hexrotor UAV and interface it with the custom designed gripper. Our framework is deployed on the specified UAV in order to report the performance analysis of the individual modules. Apart from the manipulation system, we also highlight several hidden challenges associated with the UAVs in this context.

This paper addresses the problem of cooperative transport of a point mass hoisted by two aerial robots. Treating the robots as a leader and a follower, the follower stabilizes the system with respect to the leader using only feedback from its Inertial Measurement Units (IMU). This is accomplished by neglecting the acceleration of the leader, analyzing the system through the generalized coordinates or the cables’ angles, and employing an observation model based on the IMU measurements. A lightweight estimator based on an Extended Kalman Filter (EKF) and a controller are derived to stabilize the robot-payload-robot system. The proposed methods are verified with extensive flight experiments, first with a single robot and then with two robots. The results show that the follower is capable of realizing the desired quasi-static trajectory using only its IMU measurements. The outcomes demonstrate promising progress towards the goal of autonomous cooperative transport of a suspended payload via small flying robots with minimal sensing and computational requirements.

This paper proposes a magnetic needle steering controller to manipulate mesoscale magnetic suture needles for executing planned suturing motion. This is an initial step towards our research objective: enabling autonomous control of magnetic suture needles for suturing tasks in minimally invasive surgery. To demonstrate the feasibility of accurate motion control, we employ a cardinally-arranged four-coil electromagnetic system setup and control magnetic suture needles in a 2-dimensional environment, i.e., a Petri dish filled with viscous liquid. Different from only using magnetic field gradients to control small magnetic agents under high damping conditions, the dynamics of a magnetic suture needle are investigated and encoded in the controller. Based on mathematical formulations of magnetic force and torque applied on the needle, we develop a kinematically constrained dynamic model that controls the needle to rotate and only translate along its central axis for mimicking the behavior of surgical sutures. A current controller of the electromagnetic system combining with closed-loop control schemes is designed for commanding the magnetic suture needles to achieve desired linear and angular velocities. To evaluate control performance of magnetic suture needles, we conduct experiments including needle rotation control, needle position control by using discretized trajectories, and velocity control by using a time-varying circular trajectory. The experiment results demonstrate our proposed needle steering controller can perform accurate motion control of mesoscale magnetic suture needles.

In recent years, self-supervised methods for monocular depth estimation has rapidly become an significant branch of depth estimation task, especially for autonomous driving applications. Despite the high overall precision achieved, current methods still suffer from a) imprecise object-level depth inference and b) uncertain scale factor. The former problem would cause texture copy or provide inaccurate object boundary, and the latter would require current methods to have an additional sensor like LiDAR to provide depth ground-truth or stereo camera as additional training inputs, which makes them difficult to implement. In this work, we propose to address these two problems together by introducing DNet. Our contributions are twofold: a) a novel dense connected prediction (DCP) layer is proposed to provide better object-level depth estimation and b) specifically for autonomous driving scenarios, dense geometrical constrains (DGC) is introduced so that precise scale factor can be recovered without additional cost for autonomous vehicles. Extensive experiments have been conducted and, both DCP layer and DGC module are proved to be effectively solving the aforementioned problems respectively. Thanks to DCP layer, object boundary can now be better distinguished in the depth map and the depth is more continues on object level. It is also demonstrated that the performance of using DGC to perform scale recovery is comparable to that using ground-truth information, when the camera height is given and the ground point takes up more than 1.03% of the pixels. Code is available at https://github.com/TJ-IPLab/DNet.

Currently, drone research and development has received significant attention worldwide. Particularly, delivery services employ drones as it is a viable method to improve delivery efficiency by using a several unmanned drones. Research has been conducted to realize complete automation of drone control for such services. However, regarding the takeoff and landing port of the drones, conventional methods have focused on the landing operation of a single drone, and the continuous landing of multiple drones has not been realized. To address this issue, we propose a completely novel port system, "EAGLES Port," that allows several drones to continuously land and takeoff in a short time. Experiments verified that the landing time efficiency of the proposed port is ideally 7.5 times higher than that of conventional vertical landing systems. Moreover, the system can tolerate 270 mm of horizontal positional error, ±30° of angular error in the drone’s approach (±40° with the proposed gate mechanism), and up to 1.9 m/s of drone’s approach speed. This technology significantly contributes to the scalability of drone usage. Therefore, it is critical for the development of a future drone port for the landing of automated drone swarms.

In this paper, we present an analytical modeling approach to address the problem of tension loss in a generic variable curvature tendon-driven continuum manipulators (TD-CM) occurring due to the tendon-sheath distributed friction force. Despite the previous approaches in the literature, our presented model and the iterative solution algorithm do not rely on a priori known curvature/shape of the TD-CM and can be implemented on any TD-CM with constant/ variable curvatures with a continuous neutral axis function. The performance of the proposed modeling approach in predicting the distributed tendon tension and tension loss has been evaluated via simulation and experimental studies on a TD-CM with planar bending. Results demonstrate the outstanding and accurate performance of our novel modeling and the proposed solution algorithm.

Disturbance estimation for Micro Aerial Vehicles (MAVs) is crucial for robustness and safety. In this paper, we use novel, bio-inspired airflow sensors to measure the airflow acting on a MAV, and we fuse this information in an Unscented Kalman filter (UKF) to simultaneously estimate the three-dimensional wind vector, the drag force, and other interaction forces (e.g. due to collisions, interaction with a human) acting on the robot. To this end, we present and compare a fully model-based and a deep learning-based strategy. The model-based approach considers the MAV and airflow sensor dynamics and its interaction with the wind, while the deep learning-based strategy uses a Long Short-Term Memory (LSTM) to obtain an estimate of the relative airflow, which is then fused in the proposed filter. We validate our methods in hardware experiments, showing that we can accurately estimate relative airflow of up to 4 m/s, and we can differentiate drag and interaction force.

In this paper, we present a novel algorithm for constructing a maximally informative path for a robot in an information gathering task. We use a Self-Organizing Map (SOM) framework to discover important topological features in the information function. Using these features, we identify a set of distinct classes of trajectories, each of which has improved convexity compared with the original function. We then leverage a Stochastic Gradient Ascent (SGA) optimization algorithm within each of these classes to optimize promising representative paths. The increased convexity leads to an improved chance of SGA finding the globally optimal path across all homotopy classes. We demonstrate our approach in three different simulated experiments. First, we show that our SOM is able to correctly learn the topological features of a gyre environment with a well-defined topology. Then, in the second set of experiments, we compare the effectiveness of our algorithm in an information gathering task across the gyre world, a set of randomly generated worlds, and a set of worlds drawn from real-world ocean model data. In these experiments our algorithm performs competitively or better than a state-of-the-art Branch and Bound while requiring significantly less computation time. Lastly, the final set of experiments show that our method scales better than the comparison methods across different planning mission sizes in real-world environments.

When executing a certain task, human beings can choose or make an appropriate tool to achieve the task. This research especially addresses the optimization of tool shape for robotic tool-use. We propose a method in which a robot obtains an optimized tool shape, tool trajectory, or both, depending on a given task. The feature of our method is that a transition of the task state when the robot moves a certain tool along a certain trajectory is represented by a deep neural network. We applied this method to object manipulation tasks on a 2D plane, and verified that appropriate tool shapes are generated by using this novel method.

In this paper we investigate how social robots can efficiently gather user preferences without exceeding the allowed user annoyance threshold. To do so, we use a Gazebo based simulated office environment with a TIAGo Steel robot. We then formulate the user annoyance aware preference elicitation problem as a combination of tensor completion and knapsack problems. We then test our approach on the aforementioned simulated environment and demonstrate that it can accurately estimate user preferences.

Motivated by the goal of achieving robust, drift-free pose estimation in long-term autonomous navigation, in this work we propose a methodology to fuse global positional information with visual and inertial measurements in a tightly-coupled nonlinear-optimization-based estimator. Differently from previous works, which are loosely-coupled, the use of a tightly-coupled approach allows exploiting the correlations amongst all the measurements. A sliding window of the most recent system states is estimated by minimizing a cost function that includes visual re-projection errors, relative inertial errors, and global positional residuals. We use IMU preintegration to formulate the inertial residuals and leverage the outcome of such algorithm to efficiently compute the global position residuals. The experimental results show that the proposed method achieves accurate and globally consistent estimates, with negligible increase of the optimization computational cost. Our method consistently outperforms the loosely-coupled fusion approach. The mean position error is reduced up to 50% with respect to the loosely-coupled approach in outdoor Unmanned Aerial Vehicle (UAV) flights, where the global position information is given by noisy GPS measurements. To the best of our knowledge, this is the first work where global positional measurements are tightly fused in an optimization-based visual-inertial odometry algorithm, leveraging the IMU preintegration method to define the global positional factors.

Learning computational models for visual attention (saliency estimation) is an effort to inch machines/robots closer to human visual cognitive abilities. Data-driven efforts have dominated the landscape since the introduction of deep neural network architectures. In deep learning research, the choices in architecture design are often empirical and frequently lead to more complex models than necessary. The complexity, in turn, hinders the application requirements. In this paper, we identify four key components of saliency models, i.e., input features, multi-level integration, readout architecture, and loss functions. We review the existing state of the art models on these four components and propose novel and simpler alternatives. As a result, we propose two novel end-to-end architectures called SimpleNet and MDNSal, which are neater, minimal, more interpretable and achieve state of the art performance on public saliency benchmarks. SimpleNet is an optimized encoder-decoder architecture and brings notable performance gains on the SALICON dataset (the largest saliency benchmark). MDNSal is a parametric model that directly predicts parameters of a GMM distribution and is aimed to bring more interpretability to the prediction maps. The proposed saliency models can be inferred at 25fps, making them suitable for real-time applications. Code and pre-trained models are available at https://github.com/samyak0210/saliency.

The recent progress of an ongoing project utilizing a ducted-fan propulsion system to improve a humanoid robot's ability to step over large ditches is reported. A novel method (GAS) based on the genetic algorithm with smoothness constraint can effectively minimize the thrust by optimizing the robot's posture during 3D stepping. The significant advantage of the method is that it can realize the continuity and smoothness of the thrust and pelvis trajectories. The method enables the landing point of the robot's swing foot to be not only in the forward but also in a side direction. The methods were evaluated by simulation and by being applied on a prototype robot, JetHR1. By keeping a quasistatic balance, the robot could step over a ditch with a span of 450 mm (as much as 97% of the length of the robot's leg) in 3D stepping.

The purpose of this study is to investigate a robot's impression perceived by users as well as the accuracy of perception of location information, which the robot provided according to the modality type of the robot. To explore this, we designed two 2 (verbal types: deictic vs. descriptive) x 2 (nose pointing: with nose vs. without nose) x 2 (eye pointing: with eyes vs. without eyes) mixed-participant studies. In the first study, we investigated the impacts of the robot's modality type in the imperative pointing situation. As a result, participants identified the robot's pointing gesture with nose as more effective, social, and positive, than the robot's pointing gesture without nose. Moreover, the descriptive speech robot was evaluated as more positive than the deictic speech robot. In terms of the accuracy of perception of location information, which the robot provided, participants identified the robot-designated chair more accurately when the robot delivered a deictic speech than when the robot delivered a descriptive speech. For the second study, we explored the effects of the robot's modality type in the declarative pointing situation. As a result, the robot's descriptive speech was rated as effective, social, natural, competent, trustworthy, and more positive than deictic speech. In the case of the robot's pointing gestures, pointing gesture with nose was evaluated as more effective, social, natural, competent, trustworthy, and positive than that without nose. In terms of the accuracy of location information perception, participants perceived the location of the object designated by the robot more accurately when the robot used descriptive speech, pointed with nose and without eyes.

Evidence suggests that the metabolic cost associated with the locomotive activity of walking is dependent upon ankle stiffness. This stiffness can be a control parameter in an ankle-foot prosthesis. Considering unique physical interaction between each individual with below-knee amputation and robotic ankle-foot prosthesis, individually tuned stiffness in a robotic ankle-foot prosthesis may improve assistance benefits. This personalization can be accomplished through human-in-the-loop (HIL) Bayesian optimization (BO). Here, we conducted a pilot study to identify personalized ankle- foot prosthesis stiffness using the HIL BO to minimize the cost of walking, shown by metabolic cost. We used an improved versatile ankle-foot prosthesis emulator, which enabled to test controllers with a wide range of stiffness conditions. Two participants with simulated amputation reduced their cost of walking under the condition of personalized (optimized) stiffness by 6% and 5%, respectively. This result suggests that personalized stiffness may improve assistance benefit.