ICRA 2023 Papers

This paper presents a coaxial modular aerial system (CMAS) formed by homogeneous modules driven by their center of mass. CMAS is designed to perform independent and cooperative flight with or without payload. Properties of the modularity concept allow the system to adapt to different situations and/or tasks by adding/removing modules to/from a configuration. The CMAS module is based on a coaxial motor and a two degree-of-freedom mechanism that transfers its center of mass from one side to another to make the module navigate around. The magnetic-based connector mechanism allows the module to be attached to other modules and to different metallic surfaces. A decentralized and asynchronous 3D path planning algorithm is implemented to avoid the trajectories of other modules/obstacles and ensures safe reconfiguration of the modules. Simulations within various environments show the applicability of the reconfiguration algorithm.

Point cloud registration is a crucial problem in computer vision and robotics. Existing methods either rely on matching local geometric features, which are sensitive to the pose differences, or leverage global shapes, which leads to inconsistency when facing distribution variances such as partial overlapping. Combining the advantages of both types of methods, we adopt a coarse-to-fine pipeline that concurrently handles both issues. We first reduce the pose differences between input point clouds by aligning global features; then we match the local features to further refine the inaccurate alignments resulting from distribution variances. As global feature alignment requires the features to preserve the poses of input point clouds and local feature matching expects the features to be invariant to these poses, we propose an SE(3)-equivariant feature extractor to simultaneously generate two types of features. In this feature extractor, representations that preserve the poses are first encoded by our novel SE(3)-equivariant network and then converted into pose-invariant ones by a pose-detaching module. Experiments demonstrate that our proposed method increases the recall rate by 20% compared to state-of-the-art methods when facing both pose differences and distribution variances.

Robot grasping is an actively studied area in robotics, mainly focusing on the quality of generated grasps for object manipulation. However, despite advancements, these methods do not consider the human-robot collaboration settings where robots and humans will have to grasp the same objects concurrently. Therefore, generating robot grasps compatible with human preferences of simultaneously holding an object becomes necessary to ensure a safe and natural collaboration experience. In this paper, we propose a novel, deep neural network-based method called CoGrasp that generates human-aware robot grasps by contextualizing human preference mod-els of object grasping into the robot grasp selection pro-cess. We validate our approach against existing state-of-the-art robot grasping methods through simulated and real-robot experiments and user studies. In real robot experiments, our method achieves about 88% success rate in producing stable grasps that also allow humans to interact and grasp objects simultaneously in a socially compliant manner. Furthermore, our user study with 10 independent participants indicated our approach enables a safe, natural, and socially-aware human-robot objects' co-grasping experience compared to a standard robot grasping technique.

Clothes grasping and unfolding is a core step in robotic-assisted dressing. Most existing works leverage depth images of clothes to train a deep learning-based model to recognize suitable grasping points. These methods often utilize physics engines to synthesize depth images to reduce the cost of real labeled data collection. However, the natural domain gap between synthetic and real images often leads to poor performance of these methods on real data. Furthermore, these approaches often struggle in scenarios where grasping points are occluded by the clothing item itself. To address the above challenges, we propose a novel Bi-directional Fractal Cross Fusion Network (BiFCNet) for semantic segmentation, enabling recognition of graspable regions in order to provide more possibilities for grasping. Instead of using depth images only, we also utilize RGB images with rich color features as input to our network in which the Fractal Cross Fusion (FCF) module fuses RGB and depth data by considering global complex features based on fractal geometry. To reduce the cost of real data collection, we further propose a data augmentation method based on an adversarial strategy, in which the color and geometric transformations simultaneously process RGB and depth data while maintaining the label correspondence. Finally, we present a pipeline for clothes grasping and unfolding from the perspective of semantic segmentation, through the addition of a strategy for grasp point selection from segmentation regions based on clothing flatness measures, while taking into account the grasping direction. We evaluate our BiFCNet on the public dataset NYUDv2 and obtained comparable performance to current state-of-the-art models. We also deploy our model on a Baxter robot, running extensive grasping and unfolding experiments as part of our ablation studies, achieving an 84% success rate.

Automating garment manipulation is challenging due to extremely high variability in object configurations. To reduce this intrinsic variation, we introduce the task of “canonicalized-alignment” that simplifies downstream applications by reducing the possible garment configurations. This task can be considered as “cloth state funnel” that manipulates arbitrarily configured clothing items into a predefined deformable configuration (i.e. canonicalization) at an appropriate rigid pose (i.e. alignment). In the end, the cloth items will result in a compact set of structured and highly visible configurations - which are desirable for downstream manipulation skills. To enable this task, we propose a novel canonicalized-alignment objective that effectively guides learning to avoid adverse local minima during learning. Using this objective, we learn a multi-arm, multi-primitive policy that strategically chooses between dynamic flings and quasi-static pick and place actions to achieve efficient canonicalized-alignment. We evaluate this approach on a real-world ironing and folding system that relies on this learned policy as the common first step. Empirically, we demonstrate that our task-agnostic canonicalized-alignment can enable even simple manually -designed policies to work well where they were pre-viously inadequate, thus bridging the gap between automated non-deformable manufacturing and deformable manipulation.

From both an educational and research point of view, experiments on hardware are a key aspect of robotics and control. In the last decade, many open-source hardware and software frameworks for wheeled robots have been presented, mainly in the form of unicycles and car-like robots, with the goal of making robotics accessible to a wider audience and to support control systems development. Unicycles are usually small and inexpensive, and therefore facilitate experiments in a larger fleet, but they are not suited for high-speed motion. Car-like robots are more agile, but they are usually larger and more expensive, thus requiring more resources in terms of space and money. In order to bridge this gap, we present Chronos, a new car-like 1/28th scale robot with customized open-source electronics, and CRS, an open-source software framework for control and robotics. The CRS software framework includes the implementation of various state-of-the-art algorithms for control, estimation, and multi-agent coordination. With this work, we aim to provide easier access to hardware and reduce the engineering time needed to start new educational and research projects.

Soft robotic structures may play a major role in the 4th industrial revolution. Researchers have successfully demonstrated the advantages of soft robotics over traditional robots made of rigid links and joints in many application areas. Variable stiffness links (VSL) and joints (VSJ) have been investigated to achieve on-demand forces and, at the same time, be inherently safe in interactions with humans. However, a thorough characterisation of soft and rigid robotic components is still required. This paper investigates the influence of antagonistically actuated, stiffness-controllable joint-link units (JLUs) on the performance of collaborative robots (i.e. stiffness, load capacity, repetitive precision) and characterizes the difference compared with rigid units. A JLU is made of a combination of a VSL, a VSJ, and their rigid counterparts. Experimental results show that the VSL has minor differences in terms of stiffness (0.62 ∼ 0.95), output force (0.93 ∼ 0.94), and repetitive precision compared with the rigid link. For the VSJ, our results show a significant gap compared with the servo motor with regards to maximum stiffness (0.14 ∼ 0.21) and repetitive position precision (0.07 ∼ 0.25). However, similar performance on repetitive force precision and better performance on the maximum output force (1.54 ∼ 1.55 times) are demonstrated.

We consider the problem of motion and communication planning under uncertainty with limited information from a remote sensor network. Because the remote sensors are power and bandwidth limited, we use event-triggered (ET) estimation to manage communication costs. We introduce a fast and efficient sampling-based planner which computes motion plans coupled with ET communication strategies that minimize communication costs, while satisfying constraints on the probability of reaching the goal region and the point-wise probability of collision. We derive a novel method for offline propagation of the expected state distribution, and corresponding bounds on this distribution. These bounds are used to evaluate the chance constraints in the algorithm. Case studies establish the validity of our approach and demonstrate computational efficiency and asymptotic optimality of the planner.

We present an open-source Visual-Inertial-Leg Odometry (VILO) state estimation solution for legged robots, called Cerberus, which precisely estimates position on various terrains in real-time using a set of standard sensors, including stereo cameras, IMU, joint encoders, and contact sensors. In addition to estimating robot states, we perform online kinematic parameter calibration and outlier rejection to substantially reduce position drift. Hardware experiments in various indoor and outdoor environments validate that online calibration of kinematic parameters can reduce estimation drift to less than 1% during long-distance, high-speed locomotion. Our drift results are better than those of any other state estimation method using the same set of sensors reported in the literature. Moreover, our state estimator performs well even when the robot experiences large impacts and camera occlusion. The implementation of the state estimator, along with the datasets used to compute our results, is available at https://github.com/ShuoYangRobotics/Cerberus.

With the fast development of autonomous driving technologies, there is an increasing demand for high-definition (HD) maps, which provide reliable and robust prior information about the static part of the traffic environments. As one of the important elements in HD maps, road lane centerline is critical for downstream tasks, such as prediction and planning. Manually annotating centerlines for road lanes in HD maps is labor-intensive, expensive and inefficient, severely restricting the wide applications of autonomous driving systems. Previous work seldom explores the lane centerline detection problem due to the complicated topology and severe overlapping issues of lane centerlines. In this paper, we propose a novel method named CenterLineDet to detect lane centerlines for automatic HD map generation. Our CenterLineDet is trained by imitation learning and can effectively detect the graph of centerlines with vehicle-mounted sensors (i.e., six cameras and one LiDAR) through iterations. Due to the use of the DETR-like transformer network, CenterLineDet can handle complicated graph topology, such as lane intersections. The proposed approach is evaluated on the large-scale public dataset NuScenes. The superiority of our CenterLineDet is demonstrated by the comparative results. Our code, supplementary materials, and video demonstrations are available at https://tonyxuqaq.github.io/projects/CenterLineDet/.

Leveraging multi-modal fusion, especially between camera and LiDAR, has become essential for building accurate and robust 3D object detection systems for autonomous vehicles. Until recently, point decorating approaches, in which point clouds are augmented with camera features, have been the dominant approach in the field. However, these approaches fail to utilize the higher resolution images from cameras. Recent works projecting camera features to the bird's-eye-view (BEV) space for fusion have also been proposed, however they require projecting millions of pixels, most of which only contain background information. In this work, we propose a novel approach Center Feature Fusion (CFF), in which we leverage center-based detection networks in both the camera and LiDAR streams to identify relevant object locations. We then use the center-based detection to identify the locations of pixel features relevant to object locations, a small fraction of the total number in the image. These are then projected and fused in the BEV frame. On the nuScenes dataset, we outperform the LiDAR-only baseline by 4.9% mAP while fusing up to 100x fewer features than other fusion methods.

This work presents a step towards utilizing incrementally-improving symbolic perception knowledge of the robot's surroundings for provably correct reactive control synthesis applied to an autonomous driving problem. Combining abstract models of motion control and information gathering, we show that assume-guarantee specifications (a subclass of Linear Temporal Logic) can be used to define and resolve traffic rules for cautious planning. We propose a novel representation called symbolic refinement tree for perception that captures the incremental knowledge about the environment and embodies the relationships between various symbolic perception inputs. The incremental knowledge is leveraged for synthesizing verified reactive plans for the robot. The case studies demonstrate the efficacy of the proposed approach in synthesizing control inputs even in case of partially occluded environments.

Robot data collected in complex real-world scenarios are often biased due to safety concerns, human preferences, and mission or platform constraints. Consequently, robot learning from such observational data poses great challenges for accurate parameter estimation. We propose a principled causal inference framework for robots to learn the parameters of a stochastic motion model using observational data. Specifically, we leverage the de-biasing functionality of the potential-outcome causal inference framework, the Inverse Propensity Weighting (IPW), and the Doubly Robust (DR) methods, to obtain a better parameter estimation of the robot's stochastic motion model. The IPW is a re-weighting approach to ensure unbiased estimation, and the DR approach further combines any two estimators to strengthen the unbiased result even if one of these estimators is biased. We then develop an approximate policy iteration algorithm using the bias-eliminated estimated state transition function. We validate our framework using both simulation and real-world experiments, and the results have revealed that the proposed causal inference-based navigation and control framework can correctly and efficiently learn the parameters from biased observational data.

Accurately estimating the shape of objects in dense clutters makes important contribution to robotic packing, because the optimal object arrangement requires the robot planner to acquire shape information of all existed objects. However, the objects for packing are usually piled in dense clutters with severe occlusion, and the object shape varies significantly across different instances for the same category. They respectively cause large object segmentation errors and inaccurate shape recovery on unseen instances, which both degrade the performance of shape estimation during deployment. In this paper, we propose a category-level shape estimation method for densely cluttered objects. Our framework partitions each object in the clutter via the multi-view visual information fusion to achieve high segmentation accuracy, and the instance shape is recovered by deforming the category templates with diverse geometric transformations to obtain strengthened generalization ability. Specifically, we first collect the multi-view RGB-D images of the object clutters for point cloud reconstruction. Then we fuse the feature maps representing the visual information of multi-view RGB images and the pixel affinity learned from the clutter point cloud, where the acquired instance segmentation masks of multi-view RGB images are projected to partition the clutter point cloud. Finally, the instance geometry information is obtained from the partially observed instance point cloud and the corresponding category template, and the deformation parameters regarding the template are predicted for shape estimation. Experiments in the simulated environment and real world show that our method achieves high shape estimation accuracy for densely cluttered everyday objects with various shapes.

Correspondence search is an essential step in rigid point cloud registration algorithms. Most methods maintain a single correspondence at each step and gradually remove wrong correspondances. However, building one-to-one correspondence with hard assignments is extremely difficult, especially when matching two point clouds with many locally similar features. This paper proposes an optimization method that retains all possible correspondences for each keypoint when matching a partial point cloud to a complete point cloud. These uncertain correspondences are then gradually updated with the estimated rigid transformation by considering the matching cost. More-over, we propose a new point feature descriptor that measures the similarity between local point cloud regions. Extensive experiments show that our method outperforms the state-of-the-art (SoTA) methods even when matching different objects within the same category. Notably, our method outperforms the SoTA methods when registering real-world noisy depth images to a template shape by up to 20% performance.

This manuscript introduces an object deformability-agnostic framework for co-carrying tasks that are shared between a person and multiple robots. Our approach allows the full control of the co-carrying trajectories by the person while sharing the load with multiple robots depending on the size and the weight of the object. This is achieved by merging the haptic information transferred through the object and the human motion information obtained from a motion capture system. One important advantage of the framework is that no strict internal communication is required between the robots, regardless of the object size and deformation characteristics. We validate the framework with two challenging real-world scenarios: co-transportation of a wooden rigid closet and a bulky box on top of forklift moving straps, with the latter characterizing deformable objects. In order to evaluate the generalizability of the proposed framework, a heterogenous team of two mobile manipulators that consist of an Omni-directional mobile base and a collaborative robotic arm with different DoFs is chosen for the experiments. The qualitative comparison between our controller and the baseline controller (i.e., an admittance controller) during these experiments demonstrated the effectiveness of the proposed framework especially when co-carrying deformable objects. Furthermore, we believe that the performance of our framework during the experiment with the lifting straps offers a promising solution for the co-transportation of bulky and ungraspable objects.

After many researchers observed fruitfulness from the recent diffusion probabilistic model, its effectiveness in image generation is actively studied these days. In this paper, our objective is to evaluate the potential of diffusion probabilistic models for 3D human motion-related tasks. To this end, this pa-per presents a study of employing diffusion probabilistic models to predict future 3D human motion(s) from the previously observed motion. Based on the Human 3.6M and HumanEva-I datasets, our results show that diffusion probabilistic models are competitive for both single (deterministic) and multiple (stochastic) 3D motion prediction tasks, after finishing a single training process. In addition, we find out that diffusion probabilistic models can offer an attractive compromise, since they can strike the right balance between the likelihood and diversity of the predicted future motions. Our code is publicly available on the project website: https://sites.google.com/view/diffusion-motion-prediction.

The AlphaGarden is an automated testbed for indoor polyculture farming which combines a first-order plant simulator, a gantry robot, a seed planting algorithm, plant phenotyping and tracking algorithms, irrigation sensors and algorithms, and custom pruning tools and algorithms. In this paper, we systematically compare the performance of the AlphaGarden to professional horticulturalists on the staff of the UC Berkeley Oxford Tract Greenhouse. The humans and the machine tend side-by-side polyculture gardens with the same seed arrangement. We compare performance in terms of canopy coverage, plant diversity, and water consumption. Results from two 60-day cycles suggest that the automated AlphaGarden performs comparably to professional horticulturalists in terms of coverage and diversity, and reduces water consumption by as much as 44%. Code, videos, and datasets are available at https//sites.google.com/berkeley.edulsystematiccomparison

Ultra-Wideband (UWB) systems are becoming increasingly popular for indoor localization, where range measurements are obtained by measuring the time-of-flight of radio signals. However, the range measurements typically suffer from a systematic error or bias that must be corrected for high-accuracy localization. In this paper, a ranging protocol is proposed alongside a robust and scalable antenna-delay calibration procedure to accurately and efficiently calibrate antenna delays for many UWB tags. Additionally, the bias and uncertainty of the measurements are modelled as a function of the received-signal power. The full calibration procedure is presented using experimental training data of 3 aerial robots fitted with 2 UWB tags each, and then evaluated on 2 test experiments. A localization problem is then formulated on the experimental test data, and the calibrated measurements and their modelled uncertainty are fed into an extended Kalman filter (EKF). The proposed calibration is shown to yield an average of 46% improvement in localization accuracy. Lastly, the paper is accompanied by an open-source UWB-calibration Python library, which can be found at https://github.com/decargroup/uwb_calibration.