Robotic manipulation is a key enabler for automation in the fulfillment logistics sector. Such robotic systems require perception and manipulation capabilities to handle a wide variety of objects. Existing systems either operate on a closed set of objects or perform object-agnostic manipulation which lacks the capability for deliberate and reliable manipulation at scale. Object identification (ID) unlocks the ability for large-scale, object-centric manipulation by mapping object segments to one of the previously seen objects from a database. Nevertheless, it is often limited by the availability of reference data or coverage for objects in a database. In this work, we propose to perform object identification with multiple reference databases, including images and text references, each with a different coverage and matching challenge. We propose a training strategy that tackles the challenges of learning domain-invariant image embeddings, image-text matching and fusing predictions from different sources. We perform experiments over a recent benchmark with over 190K+ unique objects, extend the dataset with the additional reference sources and propose an evaluation strategy that simulates coverage for different reference sources. Model trained with the proposed learning pipeline shows robust performance over a range of simulation experiments.

The large-scale deployment of robotic manipulation systems in warehouses has highlighted the rare but costly problem of robot-induced object damage. We present a system that uses a classification model to predict whether an object will get damaged during robotic manipulation. The model uses object attributes retrieved from warehouse information systems as well as attributes available at our robotic workcell. We evaluated different classical machine learning models, as well as a large language model (BERT) and a multimodal-transformer for our task. We show that the multi-modal transformer model that is able to leverage text and image data outperforms models that only rely on categorical and numerical data. Furthermore, our comparative analysis equips the selection the optimal model for an application. We validate our system during an experiment in which the output of the damage prediction system is used to avoid picking objects that are likely to get damaged. In over 50k pick-and-place activities, our system reduces damage rate by 64%.

This paper introduces Amazon Robotic Manipulation Benchmark (ARMBench), a large-scale, object-centric benchmark dataset for robotic manipulation in the context of a warehouse. Automation of operations in modern warehouses requires a robotic manipulator to deal with a wide variety of objects, unstructured storage, and dynamically changing inventory. Such settings pose challenges in perceiving the identity, physical characteristics, and state of objects during manipulation. Existing datasets for robotic manipulation consider a limited set of objects or utilize 3D models to generate synthetic scenes with limitation in capturing the variety of object properties, clutter, and interactions. We present a large-scale dataset collected in an Amazon warehouse using a robotic manipulator performing object singulation from containers with heterogeneous contents. ARMBench contains images, videos, and metadata that corresponds to 235K+ pick-and-place activities on 190K+ unique objects. The data is captured at different stages of manipulation, i.e., pre-pick, during transfer, and after placement. Benchmark tasks are proposed by virtue of high-quality annotations and baseline performance evaluation are presented on three visual perception challenges, namely 1) object segmentation in clutter, 2) object identification, and 3) defect detection. ARMBench can be accessed at http://armbench.com

This work proposes a robotic pipeline for picking and constrained placement of objects without geometric shape priors. Compared to recent efforts developed for similar tasks, where every object was assumed to be novel, the proposed system recognizes previously manipulated objects and per-forms online model reconstruction and reuse. Over a lifelong manipulation process, the system keeps learning features of objects it has interacted with and updates their reconstructed models. Whenever an instance of a previously manipulated object reappears, the system aims to first recognize it and then register its previously reconstructed model given the current observation. This step greatly reduces object shape uncertainty allowing the system to even reason for parts of objects, which are currently not observable. This also results in better manipulation efficiency as it reduces the need for active perception of the target object during manipulation. To get a reusable reconstructed model, the proposed pipeline adopts: i) TSDF for object representation, and ii) a variant of the standard particle filter algorithm for pose estimation and tracking of the partial object model. Furthermore, an effective way to construct and maintain a dataset of manipulated objects is presented. A sequence of real-world manipulation experiments is performed. They show how future manipulation tasks become more effective and efficient by reusing reconstructed models of previously manipulated objects, which were generated during their prior manipulation, instead of treating objects as novel every time.

Collaborative robotics requires effective communication between a robot and a human partner. This work proposes a set of interpretive principles for how a robotic arm can use pointing actions to communicate task information to people by extending existing models from the related literature. These principles are evaluated through studies where English-speaking human subjects view animations of simulated robots instructing pick-and-place tasks. The evaluation distinguishes two classes of pointing actions that arise in pick-and-place tasks: referential pointing (identifying objects) and locating pointing (identifying locations). The study indicates that human subjects show greater flexibility in interpreting the intent of referential pointing compared to locating pointing, which needs to be more deliberate. The results also demonstrate the effects of variation in the environment and task context on the interpretation of pointing. Our corpus, experiments and design principles advance models of context, common sense reasoning and communication in embodied communication.

Many manipulation tasks, such as placement or within-hand manipulation, require the object's pose relative to a robot hand. The task is difficult when the hand significantly occludes the object. It is especially hard for adaptive hands, for which it is not easy to detect the finger's configuration. In addition, RGB-only approaches face issues with texture-less objects or when the hand and the object look similar. This paper presents a depth-based framework, which aims for robust pose estimation and short response times. The approach detects the adaptive hand's state via efficient parallel search given the highest overlap between the hand's model and the point cloud. The hand's point cloud is pruned and robust global registration is performed to generate object pose hypotheses, which are clustered. False hypotheses are pruned via physical reasoning. The remaining poses' quality is evaluated given agreement with observed data. Extensive evaluation on synthetic and real data demonstrates the accuracy and computational efficiency of the framework when applied on challenging, highly-occluded scenarios for different object types. An ablation study identifies how the framework's components help in performance. This work also provides a dataset for in-hand 6D object pose estimation. Code and dataset are available at: https://github.com/wenbowen123/icra20-hand-object-pose.

Picking an item in the presence of other objects can be challenging as it involves occlusions and partial views. Given object models, one approach is to perform object pose estimation and use the most likely candidate pose per object to pick the target without collisions. This approach, however, ignores the uncertainty of the perception process both regarding the target's and the surrounding objects' poses. This work proposes first a perception process for 6D pose estimation, which returns a discrete distribution of object poses in a scene. Then, an open-loop planning pipeline is proposed to return safe and effective solutions for moving a robotic arm to pick, which (a) minimizes the probability of collision with the obstructing objects; and (b) maximizes the probability of reaching the target item. The planning framework models the challenge as a stochastic variant of the Minimum Constraint Removal (MCR) problem. The effectiveness of the methodology is verified given both simulated and real data in different scenarios. The experiments demonstrate the importance of considering the uncertainty of the perception process in terms of safe execution. The results also show that the methodology is more effective than conservative MCR approaches, which avoid all possible object poses regardless of the reported uncertainty.

Tracking the 6D pose of objects in video sequences is important for robot manipulation. This task, however, introduces multiple challenges: (i) robot manipulation involves significant occlusions; (ii) data and annotations are troublesome and difficult to collect for 6D poses, which complicates machine learning solutions, and (iii) incremental error drift often accumulates in long term tracking to necessitate re-initialization of the object's pose. This work proposes a data-driven optimization approach for long-term, 6D pose tracking. It aims to identify the optimal relative pose given the current RGB-D observation and a synthetic image conditioned on the previous best estimate and the object's model. The key contribution in this context is a novel neural network architecture, which appropriately disentangles the feature encoding to help reduce domain shift, and an effective 3D orientation representation via Lie Algebra. Consequently, even when the network is trained only with synthetic data can work effectively over real images. Comprehensive experiments over benchmarks - existing ones as well as a new dataset with significant occlusions related to object manipulation - show that the proposed approach achieves consistently robust estimates and outperforms alternatives, even though they have been trained with real images. The approach is also the most computationally efficient among the alternatives and achieves a tracking frequency of 90.9Hz.

Accurate pose estimation is often a requirement for robust robotic grasping and manipulation of objects placed in cluttered, tight environments, such as a shelf with multiple objects. When deep learning approaches are employed to perform this task, they typically require a large amount of training data. However, obtaining precise 6 degrees of freedom for ground-truth can be prohibitively expensive. This work therefore proposes an architecture and a training process to solve this issue. More precisely, we present a weak object detector that enables localizing objects and estimating their 6D poses in cluttered and occluded scenes. To minimize the human labor required for annotations, the proposed detector is trained with a combination of synthetic and a few weakly annotated real images (as little as 10 images per object), for which a human provides only a list of objects present in each image (no time-consuming annotations, such as bounding boxes, segmentation masks and object poses). To close the gap between real and synthetic images, we use multiple domain classifiers trained adversarially. During the inference phase, the resulting class-specific heatmaps of the weak detector are used to guide the search of 6D poses of objects. Our proposed approach is evaluated on several publicly available datasets for pose estimation. We also evaluated our model on classification and localization in unsupervised and semi-supervised settings. The results clearly indicate that this approach could provide an efficient way toward fully automating the training process of computer vision models used in robotics.

Advances in sensor technologies, object detection algorithms, planning frameworks and hardware designs have motivated the deployment of robots in warehouse automation. A variety of such applications, like order fulfillment or packing tasks, require picking objects from unstructured piles and carefully arranging them in bins or containers. Desirable solutions need to be low-cost, easily deployable and controllable, making minimalistic hardware choices desirable. The challenge in designing an effective solution to this problem relates to appropriately integrating multiple components, so as to achieve a robust pipeline that minimizes failure conditions. The current work proposes a complete pipeline for solving such packing tasks, given access only to RGB-D data and a single robot arm with a vacuum-based end-effector, which is also used as a pushing finger. To achieve the desired level of robustness, three key manipulation primitives are identified, which take advantage of the environment and simple operations to successfully pack multiple cubic objects. The overall approach is demonstrated to be robust to execution and perception errors. The impact of each manipulation primitive is evaluated by considering different versions of the proposed pipeline, which incrementally introduce reasoning about object poses and corrective manipulation actions.

This work proposes a process for efficiently searching over combinations of individual object 6D pose hypotheses in cluttered scenes, especially in cases involving occlusions and objects resting on each other. The initial set of candidate object poses is generated from state-of-the-art object detection and global point cloud registration techniques. The best scored pose per object by using these techniques may not be accurate due to overlaps and occlusions. Nevertheless, experimental indications provided in this work show that object poses with lower ranks may be closer to the real poses than ones with high ranks according to registration techniques. This motivates a global optimization process for improving these poses by taking into account scene-level physical interactions between objects. It also implies that the Cartesian product of candidate poses for interacting objects must be searched so as to identify the best scene-level hypothesis. To perform the search efficiently, the candidate poses for each object are clustered so as to reduce their number but still keep a sufficient diversity. Then, searching over the combinations of candidate object poses is performed through a Monte Carlo Tree Search (MCTS) process that uses the similarity between the observed depth image of the scene and a rendering of the scene given the hypothesized pose as a score that guides the search procedure. MCTS handles in a principled way the tradeoff between fine-tuning the most promising poses and exploring new ones, by using the Upper Confidence Bound (UCB) technique. Experimental results indicate that this process is able to quickly identify in cluttered scenes physically-consistent object poses that are significantly closer to ground truth compared to poses found by point cloud registration methods.

Progress has been achieved recently in object detection given advancements in deep learning. Nevertheless, such tools typically require a large amount of training data and significant manual effort to label objects. This limits their applicability in robotics, where solutions must scale to a large number of objects and variety of conditions. This work proposes an autonomous process for training a Convolutional Neural Network (CNN) for object detection and pose estimation in robotic setups. The focus is on detecting objects placed in cluttered, tight environments, such as a shelf with multiple objects. In particular, given access to 3D object models, several aspects of the environment are physically simulated. The models are placed in physically realistic poses with respect to their environment to generate a labeled synthetic dataset. To further improve object detection, the network self-trains over real images that are labeled using a robust multi-view pose estimation process. The proposed training process is evaluated on several existing datasets and on a dataset collected for this paper with a Motoman robotic arm. Results show that the proposed approach outperforms popular training processes relying on synthetic - but not physically realistic - data and manual annotation. The key contributions are the incorporation of physical reasoning in the synthetic data generation process and the automation of the annotation process over real images.