Robots operating in the real world face significant but unavoidable issues in object localization that must be dealt with. A typical approach to address this is the addition of compliance mechanisms to hardware to absorb and compensate for some of these errors. However, for fine-grained manipulation tasks, the location and choice of appropriate compliance mechanisms are critical for success. For objects to be inserted in a target site on a flat surface, the object must first be successfully aligned with the opening of the slot, as well as correctly oriented along its central axis, before it can be inserted. We developed the Four-Axis Adaptive Finger Hand (FAAF hand) that is equipped with fingers that can passively adapt in four axes (x, y, z, yaw) enabling it to perform insertion tasks including lid fitting in the presence of significant localization errors. Furthermore, this adaptivity allows the use of simple control methods without requiring contact sensors or other devices. Our results confirm the ability of the FAAF hand on challenging insertion tasks of square and triangle-shaped pegs (or prisms) and placing of container lids in the presence of position errors in all directions and rotational error along the object’s central axis, using a simple control scheme.

Micro well-plates are an apparatus commonly used in chemical and biological experiments that are a few centimeters thick and contain wells or divets. In this paper, we aim to solve the task of placing the well-plate onto a well-plate holder (referred to as holder). This task is challenging due to the holder’s raised grooves being a few millimeters in height, with a clearance of less than 1 mm between the well-plate and holder, thus requiring precise control during placing. Our placing task has the following challenges: 1) The holder’s detected pose is uncertain; 2) the required accuracy is at the millimeter to sub-millimeter level due to the raised groove’s shallow height and small clearance; 3) the holder is not fixed to a desk and is susceptible to movement from external forces. To address these challenges, we developed methods including a) using tactile sensors for accurate pose estimation of the grasped well-plate to handle issue (1); b) sliding the well-plate onto the target holder while maintaining contact with the holder’s groove and estimating its orientation for accurate alignment. This allows for high precision control (addressing issue (2)) and prevents displacement of the holder during placement (addressing issue (3)). We demonstrate a high success rate for the well-plate placing task, even under noisy observation of the holder’s pose.4

Ensuring stable object placement is crucial to prevent objects from toppling over, breaking, or causing spills. When an object makes initial contact to a surface, and some force is exerted, the moment of rotation caused by the instability of the object’s placing can cause the object to rotate in a certain direction (henceforth referred to as direction of corrective rotation). Existing methods often employ a Force/Torque (F/T) sensor to estimate the direction of corrective rotation by detecting the moment of rotation as a torque. However, its effectiveness may be hampered by sensor noise and the tension of the external wiring of robot cables. To address these issues, we propose a method for stable object placing using GelSights, vision-based tactile sensors, as an alternative to F/T sensors. Our method estimates the direction of corrective rotation of objects using the displacement of the black dot pattern on the elastomeric surface of GelSight. We calculate the Curl from vector analysis, indicative of the rotational field magnitude and direction of the displacement of the black dots pattern. Simultaneously, we calculate the difference (Diff) of displacement between the left and right fingers’ GelSight’s black dots. Then, the robot can manipulate the objects’ pose using Curl and Diff features, facilitating stable placing. Across experiments, handling 18 differently characterized objects, our method achieves precise placing accuracy (less than 1-degree error) in nearly 100% of cases.

Acquiring accurate depth information of transparent objects using off-the-shelf RGB-D cameras is a well-known challenge in Computer Vision and Robotics. Depth estimation/completion methods are typically employed and trained on datasets with quality depth labels acquired from either simulation, additional sensors or specialized data collection setups and known 3d models. However, acquiring reliable depth information for datasets at scale is not straightforward, limiting training scalability and generalization. Neural Radiance Fields (NeRFs) are learning-free approaches and have demonstrated wide success in novel view synthesis and shape recovery. However, heuristics and controlled environments (lights, backgrounds, etc) are often required to accurately capture specular surfaces. In this paper, we propose using Visual Foundation Models (VFMs) for segmentation in a zero-shot, label-free way to guide the NeRF reconstruction process for these objects via the simultaneous reconstruction of semantic fields and extensions to increase robustness. Our proposed method Segmentation-AIDed NeRF (SAID-NeRF) shows significant performance on depth completion datasets for transparent objects and robotic grasping.

To perform dynamic cable manipulation to realize the configuration specified by a target image, we formulate dynamic cable manipulation as a stochastic forward model. Then, we propose a method to handle uncertainty by maximizing the expectation, which also considers estimation errors of the trained model. To avoid issues like multiple local minima and requirement of differentiability by gradient-based methods, we propose using a black-box optimization (BBO) to optimize joint angles to realize a goal image. Among BBO, we use the Tree-structured Parzen Estimator (TPE), a type of Bayesian optimization. By incorporating constraints into the TPE, the optimized joint angles are constrained within the range of motion. Since TPE is population-based, it is better able to detect multiple feasible configurations using the estimated inverse model. We evaluated image similarity between the target and cable images captured by executing the robot using optimal transport distance. The results show that the proposed method improves accuracy compared to conventional gradient-based approaches and methods that use deterministic models that do not consider uncertainty.

A problem that plagues robotic grasping is the misalignment of the object and gripper due to difficulties in precise localization, actuation, etc. Under-actuated robotic hands with compliant mechanisms are used to adapt and compensate for these inaccuracies. However, these mechanisms come at the cost of controllability and coordination. For instance, adaptive functions that let the fingers of a two-fingered gripper adapt independently may affect the coordination necessary for grasping small objects. In this work, we develop a two-fingered robotic hand capable of grasping objects that are offset from the gripper's center, while still having the requisite coordination for grasping small objects via a novel gear-type synchronization mechanism with a magnet. This gear synchronization mechanism allows the adaptive finger's tips to be aligned enabling it to grasp objects as small as toothpicks and washers. The magnetic component allows this coordination to automatically turn off when needed, allowing for the grasping of objects that are offset/misaligned from the gripper. This equips the hand with the capability of grasping light, fragile objects (strawberries, creampuffs, etc.) to heavy frying pan lids, all while maintaining their position and posture which is vital in numerous applications that require precise positioning or careful manipulation.

The food packaging industry handles an immense variety of food products with wide-ranging shapes and sizes, even within one kind of food. Menus are also diverse and change frequently, making automation of pick-and-place difficult. A popular approach to bin-picking is to first identify each piece of food in the tray by using an instance segmentation method. However, human annotations to train these methods are unreli-able and error-prone since foods are packed close together with unclear boundaries and visual similarity making separation of pieces difficult. To address this problem, we propose a method that trains purely on synthetic data and successfully transfers to the real world using sim2real methods by creating datasets of filled food trays using high-quality 3d models of real pieces of food for the training instance segmentation models. Another concern is that foods are easily damaged during grasping. We address this by introducing two additional methods- a novel adaptive finger mechanism to passively retract when a collision occurs, and a method to filter grasps that are likely to cause damage to neighbouring pieces of food during a grasp. We demonstrate the effectiveness of the proposed method on several kinds of real foods.

Food packing industry workers typically pick a target amount of food by hand from a food tray and place them in containers. Since menus are diverse and change frequently, robots must adapt and learn to handle new foods in a short time-span. Learning to grasp a specific amount of granular food requires a large training dataset, which is challenging to collect reasonably quickly. In this study, we propose ways to reduce the necessary amount of training data by augmenting a deep neural network with models that estimate its uncertainty through self-supervised learning. To further reduce human effort, we devise a data collection system that automatically generates labels. We build on the idea that we can grasp sufficiently well if there is at least one low-uncertainty (high-confidence) grasp point among the various grasp point candidates. We evaluate the methods we propose in this work on a variety of granular foods-coffee beans, rice, oatmeal and peanuts, each of which has a different size, shape and material properties such as volumetric mass density or friction. For these foods, we show significantly improved grasp accuracy of user-specified target masses using smaller datasets by incorporating uncertainty.

Estimation of tactile properties from vision, such as slipperiness or roughness, is important to effectively interact with the environment. These tactile properties help humans, as well as robots, decide which actions they should choose and how to perform them. We, therefore, propose a model to estimate the degree of tactile properties from visual perception alone (e.g., the level of slipperiness or roughness). Our method extends an encoder-decoder network, in which the latent variables are visual and tactile features. In contrast to previous works, our method does not require manual labeling, but only RGB images and the corresponding tactile sensor data. All our data is collected with a webcam and tactile sensor mounted on the end-effector of a robot, which strokes the material surfaces. We show that our model generalizes to materials not included in the training data.

For in-hand manipulation, estimation of the object pose inside the hand is one of the important functions to manipulate objects to the target pose. Since in-hand manipulation tends to cause occlusions by the hand or the object itself, image information only is not sufficient for in-hand object pose estimation. Multiple modalities can be used in this case, the advantage is that other modalities can compensate for occlusion, noise, and sensor malfunctions. Even though deciding the utilization rate of a modality (referred to as reliability value) corresponding to the situations is important, the manual design of such models is difficult, especially for various situations. In this paper, we propose deep gated multi-modal learning, which self-determines the reliability value of each modality through end-to-end deep learning. For the experiments, an RGB camera and a GelSight tactile sensor were attached to the parallel gripper of the Sawyer robot, and the object pose changes were estimated during grasping. A total of 15 objects were used in the experiments. In the proposed model, the reliability values of the modalities were determined according to the noise level and failure of each modality, and it was confirmed that the pose change was estimated even for unknown objects.

We propose a method to annotate segmentation masks accurately and automatically using invisible marker for object manipulation. Invisible marker is invisible under visible (regular) light conditions, but becomes visible under invisible light, such as ultraviolet (UV) light. By painting objects with the invisible marker, and by capturing images while alternately switching between regular and UV light at high speed, massive annotated datasets are created quickly and inexpensively. We show a comparison between our proposed method and manual annotations. We demonstrate semantic segmentation for deformable objects including clothes, liquids, and powders under controlled environmental light conditions. In addition, we show demonstrations of liquid pouring tasks under uncontrolled environmental light conditions in complex environments such as inside the office, house, and outdoors. Furthermore, it is possible to capture data while the camera is in motion so it becomes easier to capture large datasets, as shown in our demonstration.

Estimation of tactile properties from vision, such as slipperiness or roughness, is important to effectively interact with the environment. These tactile properties help us decide which actions we should choose and how to perform them. E.g., we can drive slower if we see that we have bad traction or grasp tighter if an item looks slippery. We believe that this ability also helps robots to enhance their understanding of the environment, and thus enables them to tailor their actions to the situation at hand. We therefore propose a model to estimate the degree of tactile properties from visual perception alone (e.g., the level of slipperiness or roughness). Our method extends a encoder-decoder network, in which the latent variables are visual and tactile features. In contrast to previous works, our method does not require manual labeling, but only RGB images and the corresponding tactile sensor data. All our data is collected with a webcam and uSkin tactile sensor mounted on the end-effector of a Sawyer robot, which strokes the surfaces of 25 different materials. We show that our model generalizes to materials not included in the training data by evaluating the feature space, indicating that it has learned to associate important tactile properties with images.

Comprehension of spoken natural language is an essential skill for robots to communicate with humans effectively. However, handling unconstrained spoken instructions is challenging due to (1) complex structures and the wide variety of expressions used in spoken language, and (2) inherent ambiguity of human instructions. In this paper, we propose the first comprehensive system for controlling robots with unconstrained spoken language, which is able to effectively resolve ambiguity in spoken instructions. Specifically, we integrate deep learning-based object detection together with natural language processing technologies to handle unconstrained spoken instructions, and propose a method for robots to resolve instruction ambiguity through dialogue. Through our experiments on both a simulated environment as well as a physical industrial robot arm, we demonstrate the ability of our system to understand natural instructions from human operators effectively, and show how higher success rates of the object picking task can be achieved through an interactive clarification process.

In this study, we propose a neural network based model for learning a robot's drawing sequences in an unsupervised manner. We focus on the ability to learn visual-motor relationships, which can work as a reusable memory in association of drawing motion from a picture image. Assuming that a humanoid robot can draw a shape on a pen tablet, the proposed model learns drawing sequences, which comprises drawing motion and drawn picture image frames. To learn raw pixel data without any given specific features, we utilized a deep neural network for compressing large dimensional picture images and a continuous time recurrent neural network for integration of motion and picture images. To confirm the ability of the proposed model, we performed an experiment for learning 15 sequences comprising three types of shapes. The model successfully learns all the sequences and can associate a drawing motion from a not trained picture image and a trained picture with similar success. We also show that the proposed model self-organizes its behavior according to types shapes.

We propose a method for realizing effective dynamic motion learning in a flexible-joint robot using motor babbling. Flexible-joint robots have recently attracted attention because of their adaptiveness, safety, and, in particular, dynamic motions. It is difficult to control robots that require dynamic motion. In past studies, attractors and oscillators were designed as motion primitives of an assumed task in advance. However, it is difficult to adapt to unintended environmental changes using such methods. To overcome this problem, we use a recurrent neural network (RNN) that does not require predetermined parameters. In this research, we propose a method for facilitating effective learning. First, a robot learns simple motions via motor babbling, acquiring body dynamics using a recurrent neural network (RNN). Motor babbling is the process of movement that infants use to acquire their own body dynamics during their early days. Next, the robot learns additional motions required for a target task using the acquired body dynamics. For acquiring these body dynamics, the robot uses motor babbling with its redundant flexible joints to learn motion primitives. This redundancy implies that there are numerous possible motion patterns. In comparison to a basic learning task, the motion primitives are simply modified to adjust to the task. Next, we focus on the types of motions used in motor babbling. We classify the motions into two motion types, passive motion and active motion. Passive motion involves inertia without any torque input, whereas active motion involves a torque input. The robot acquires body dynamics from the passive motion and a means of torque generation from the active motion. As a result, we demonstrate the importance of performing prior learning via motor babbling before learning a task. In addition, task learning is made more efficient by dividing the motion into two types of motor babbling patterns.