Robots that assist in daily life are required to locate specific instances of objects that match the user's desired object in the environment. This task is known as Instance-Specific Image Goal Navigation (InstanceImageNav), which requires a model capable of distinguishing between different instances within the same class. One significant challenge in robotics is that when a robot observes the same object from various 3D viewpoints, its appearance may differ greatly, making it difficult to recognize and locate the object accurately. In this study, we introduce a method, SimView, that leverages multi-view images based on a 3D semantic map of the environment and self-supervised learning by SimSiam to train an instance identification model on-site. The effectiveness of our approach is validated using a photorealistic simulator, Habitat Matterport 3D, created by scanning real home environments. Our results demonstrate a 1.7-fold improvement in task accuracy compared to CLIP, which is pre-trained multimodal contrastive learning for object search. This improvement highlights the benefits of our proposed fine-tuning method in enhancing the performance of assistive robots in InstanceImageNav tasks. The project website is https://emergentsystemlabstudent.github.io/MultiViewRetrieve/.
Improving instance-specific image goal navigation (InstanceImageNav), which locates the identical object in a real-world environment from a query image, is essential for robotic systems to assist users in finding desired objects. The challenge lies in the domain gap between low-quality images observed by the moving robot, characterized by motion blur and low-resolution, and high-quality query images provided by the user. Such domain gaps could significantly reduce the task success rate but have not been the focus of previous work. To address this, we propose a novel method called Few-shot Cross-quality Instance-aware Adaptation (CrossIA), which employs contrastive learning with an instance classifier to align features between massive low- and few high-quality images. This approach effectively reduces the domain gap by bringing the latent representations of cross-quality images closer on an instance basis. Additionally, the system integrates an object image collection with a pre-trained deblurring model to enhance the observed image quality. Our method fine-tunes the SimSiam model, pre-trained on ImageNet, using CrossIA. We evaluated our method's effectiveness through an InstanceImageNav task with 20 different types of instances, where the robot identifies the same instance in a real-world environment as a high-quality query image. Our experiments showed that our method improves the task success rate by up to three times compared to the baseline, a conventional approach based on SuperGlue. These findings highlight the potential of leveraging contrastive learning and image enhancement techniques to bridge the domain gap and improve object localization in robotic applications. The project website is https://emergentsystemlabstudent.github.io/DomainBridgingNav/.
Large language models (LLMs) and vision-language models (VLMs) have been increasingly used in robotics for high-level cognition, but their use for low-level cognition, such as interpreting sensor information, remains underexplored. In robotic grasping, estimating the reflectance of objects is crucial for successful grasping, as it significantly impacts the distance measured by proximity sensors. We investigate whether LLMs can estimate reflectance from object names alone, leveraging the embedded human knowledge in distributional semantics, and if the latent structure of language in VLMs positively affects image-based reflectance estimation. In this paper, we verify that 1) LLMs such as GPT-3.5 and GPT-4 can estimate an object's reflectance using only text as input; and 2) VLMs such as CLIP can increase their generalization capabilities in reflectance estimation from images. Our experiments show that GPT-4 can estimate an object's reflectance using only text input with a mean error of 14.7%, lower than the image-only ResNet. Moreover, CLIP achieved the lowest mean error of 11.8%, while GPT-3.5 obtained a competitive 19.9% compared to ResNet's 17.8%. These results suggest that the distributional semantics in LLMs and VLMs increases their generalization capabilities, and the knowledge acquired by VLMs benefits from the latent structure of language.
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. An accompanying video is available at the following link: https://youtu.be/fQbmCksVHlU
In this paper, a novel approach is proposed for learning robot control in contact-rich tasks such as wiping, by developing Diffusion Contact Model (DCM). Previous methods of learning such tasks relied on impedance control with time-varying stiffness tuning by performing Bayesian optimization by trial-and-error with robots. The proposed approach aims to reduce the cost of robot operation by predicting the robot contact trajectories from the variable stiffness inputs and using neural models. However, contact dynamics are inherently highly nonlinear, and their simulation requires iterative computations such as convex optimization. Moreover, approximating such computations by using finite-layer neural models is difficult. To overcome these limitations, the proposed DCM used the denoising diffusion models that could simulate the complex dynamics via iterative computations of multi-step denoising, thus improving the prediction accuracy. Stiffness tuning experiments conducted in simulated and real environments showed that the DCM achieved comparable performance to a conventional robot-based optimization method while reducing the number of robot trials.
As a sub-discipline of evolutionary and computational linguistics, emergent communication (EC) studies communication protocols, called emergent languages, arising in simulations where agents communicate. A key goal of EC is to give rise to languages that share statistical properties with natural languages. In this paper, we reinterpret Lewis's signaling game, a frequently used setting in EC, as beta-VAE and reformulate its objective function as ELBO. Consequently, we clarify the existence of prior distributions of emergent languages and show that the choice of the priors can influence their statistical properties. Specifically, we address the properties of word lengths and segmentation, known as Zipf's law of abbreviation (ZLA) and Harris's articulation scheme (HAS), respectively. It has been reported that the emergent languages do not follow them when using the conventional objective. We experimentally demonstrate that by selecting an appropriate prior distribution, more natural segments emerge, while suggesting that the conventional one prevents the languages from following ZLA and HAS.
Self-supervised learning (SSL) using mixed images has been studied to learn various image representations. Existing methods using mixed images learn a representation by maximizing the similarity between the representation of the mixed image and the synthesized representation of the original images. However, few methods consider the synthesis of representations from the perspective of mathematical logic. In this study, we focused on a synthesis method of representations. We proposed a new SSL with mixed images and a new representation format based on many-valued logic. This format can indicate the feature-possession degree, that is, how much of each image feature is possessed by a representation. This representation format and representation synthesis by logic operation realize that the synthesized representation preserves the remarkable characteristics of the original representations. Our method performed competitively with previous representation synthesis methods for image classification tasks. We also examined the relationship between the feature-possession degree and the number of classes of images in the multilabel image classification dataset to verify that the intended learning was achieved. In addition, we discussed image retrieval, which is an application of our proposed representation format using many-valued logic.
The process of industrial box-packing, which involves the accurate placement of multiple objects, requires high-accuracy positioning and sequential actions. When a robot is tasked with placing an object at a specific location with high accuracy, it is important not only to have information about the location of the object to be placed, but also the posture of the object grasped by the robotic hand. Often, industrial box-packing requires the sequential placement of identically shaped objects into a single box. The robot's action should be determined by the same learned model. In factories, new kinds of products often appear and there is a need for a model that can easily adapt to them. Therefore, it should be easy to collect data to train the model. In this study, we designed a robotic system to automate real-world industrial tasks, employing a vision-based learning control model. We propose in-hand-view-sensitive Newtonian variational autoencoder (ihVS-NVAE), which employs an RGB camera to obtain in-hand postures of objects. We demonstrate that our model, trained for a single object-placement task, can handle sequential tasks without additional training. To evaluate efficacy of the proposed model, we employed a real robot to perform sequential industrial box-packing of multiple objects. Results showed that the proposed model achieved a 100% success rate in industrial box-packing tasks, thereby outperforming the state-of-the-art and conventional approaches, underscoring its superior effectiveness and potential in industrial tasks.
Rather than traditional position control, impedance control is preferred to ensure the safe operation of industrial robots programmed from demonstrations. However, variable stiffness learning studies have focused on task performance rather than safety (or compliance). Thus, this paper proposes a novel stiffness learning method to satisfy both task performance and compliance requirements. The proposed method optimizes the task and compliance objectives (T/C objectives) simultaneously via multi-objective Bayesian optimization. We define the stiffness search space by segmenting a demonstration into task phases, each with constant responsible stiffness. The segmentation is performed by identifying impedance control-aware switching linear dynamics (IC-SLD) from the demonstration. We also utilize the stiffness obtained by proposed IC-SLD as priors for efficient optimization. Experiments on simulated tasks and a real robot demonstrate that IC-SLD-based segmentation and the use of priors improve the optimization efficiency compared to existing baseline methods.
This paper proposes a generative probabilistic model integrating emergent communication and multi-agent reinforcement learning. The agents plan their actions by probabilistic inference, called control as inference, and communicate using messages that are latent variables and estimated based on the planned actions. Through these messages, each agent can send information about its actions and know information about the actions of another agent. Therefore, the agents change their actions according to the estimated messages to achieve cooperative tasks. This inference of messages can be considered as communication, and this procedure can be formulated by the Metropolis-Hasting naming game. Through experiments in the grid world environment, we show that the proposed PGM can infer meaningful messages to achieve the cooperative task.