Search

Uncertainty of environments has long been a difficult characteristic to handle, when performing real-world robot tasks. This is because the uncertainty produces unexpected observations that cannot be covered by manual scripting. Learning based robot controlling methods are a promising approach for generating flexible motions against unknown situations, but still tend to suffer under uncertainty due to its deterministic nature. In order to adaptively perform the target task under such conditions, the robot control model must be able to accurately understand the possible uncertainty, and to exploratively derive the optimal action that minimizes such uncertainty. This paper extended an existing predictive learning based robot control method, which employ foresight prediction using dynamic internal simulation. The foresight module refines the model's hidden states by sampling multiple possible futures and replace with the one that led to the lower future uncertainty. The adaptiveness of the model was evaluated on a door opening task. The door can be opened either by pushing, pulling, or sliding, but robot cannot visually distinguish which way, and is required to adapt on the fly. The results showed that the proposed model adaptively diverged its motion through interaction with the door, whereas conventional methods failed to stably diverge. The models were analyzed on Lyapunov exponents of RNN hidden states which reflect the possible divergence at each time step during task execution. The result indicated that the foresight module biased the model to consider future consequences, which lead to embedding uncertainties at the policy of the robot controller, rather than the resultant observation. This is beneficial for implementing adaptive behaviors, which indices derivation of diverse motion during exploration.

Achieving both high speed and precision in robot operations is a significant challenge for social implementation. While factory robots excel at predefined tasks, they struggle with environment-specific actions like cleaning and cooking. Deep learning research aims to address this by enabling robots to autonomously execute behaviors through end-to-end learning with sensor data. RT-1 and ACT are notable examples that have expanded robots' capabilities. However, issues with model inference speed and hand position accuracy persist. High-quality training data and fast, stable inference mechanisms are essential to overcome these challenges. This paper proposes a motion generation model for high-speed, high-precision tasks, exemplified by the sports stacking task. By teaching motions slowly and inferring at high speeds, the model achieved a 94% success rate in stacking cups with a real robot., Comment: 2 pages, 2 figures

Preceding a core-collapse supernova, various processes produce an increasing amount of neutrinos of all flavors characterized by mounting energies from the interior of massive stars. Among them, the electron antineutrinos are potentially detectable by terrestrial neutrino experiments such as KamLAND and Super-Kamiokande via inverse beta decay interactions. Once these pre-supernova neutrinos are observed, an early warning of the upcoming core-collapse supernova can be provided. In light of this, KamLAND and Super-Kamiokande, both located in the Kamioka mine in Japan, have been monitoring pre-supernova neutrinos since 2015 and 2021, respectively. Recently, we performed a joint study between KamLAND and Super-Kamiokande on pre-supernova neutrino detection. A pre-supernova alert system combining the KamLAND detector and the Super-Kamiokande detector was developed and put into operation, which can provide a supernova alert to the astrophysics community. Fully leveraging the complementary properties of these two detectors, the combined alert is expected to resolve a pre-supernova neutrino signal from a 15 M$_{\odot}$ star within 510 pc of the Earth, at a significance level corresponding to a false alarm rate of no more than 1 per century. For a Betelgeuse-like model with optimistic parameters, it can provide early warnings up to 12 hours in advance., Comment: Resubmitted to ApJ. 22 pages, 16 figures, for more information about the combined pre-supernova alert system, see https://www.lowbg.org/presnalarm/

Direct detection of weakly interacting massive particles (WIMPs) can provide strong evidence of their existence and the directional method would have an advantage over other methods to detect the clear signal of WIMPs. Time projection chambers with micro-patterned gaseous detectors (MPGDs) are one of the common devices used in directional WIMP searches. A micro pixel chamber ($\mu$-PIC), one of the various types of MPGDs, with specially selected low background materials (LBG$\mu$-PIC) was developed and its performance was studied. The radon emission of the LBG$\mu$-PIC was less than 1/60 of that of the $\mu$-PIC currently in use. Although a non-negligible gain non-homogeneity was seen for the LBG$\mu$-PIC, it can be used for the directional WIMP search with the correction of the non-homogeneity., Comment: 18 pages, 14 figures, submitted to NIM A

Anchor-bolt insertion is a peg-in-hole task performed in the construction field for holes in concrete. Efforts have been made to automate this task, but the variable lighting and hole surface conditions, as well as the requirements for short setup and task execution time make the automation challenging. In this study, we introduce a vision and proprioceptive data-driven robot control model for this task that is robust to challenging lighting and hole surface conditions. This model consists of a spatial attention point network (SAP) and a deep reinforcement learning (DRL) policy that are trained jointly end-to-end to control the robot. The model is trained in an offline manner, with a sample-efficient framework designed to reduce training time and minimize the reality gap when transferring the model to the physical world. Through evaluations with an industrial robot performing the task in 12 unknown holes, starting from 16 different initial positions, and under three different lighting conditions (two with misleading shadows), we demonstrate that SAP can generate relevant attention points of the image even in challenging lighting conditions. We also show that the proposed model enables task execution with higher success rate and shorter task completion time than various baselines. Due to the proposed model's high effectiveness even in severe lighting, initial positions, and hole conditions, and the offline training framework's high sample-efficiency and short training time, this approach can be easily applied to construction., Comment: Published in IEEE Robotics and Automation Letters on 08 February 2023

To support humans in their daily lives, robots are required to autonomously learn, adapt to objects and environments, and perform the appropriate actions. We tackled on the task of cooking scrambled eggs using real ingredients, in which the robot needs to perceive the states of the egg and adjust stirring movement in real time, while the egg is heated and the state changes continuously. In previous works, handling changing objects was found to be challenging because sensory information includes dynamical, both important or noisy information, and the modality which should be focused on changes every time, making it difficult to realize both perception and motion generation in real time. We propose a predictive recurrent neural network with an attention mechanism that can weigh the sensor input, distinguishing how important and reliable each modality is, that realize quick and efficient perception and motion generation. The model is trained with learning from the demonstration, and allows the robot to acquire human-like skills. We validated the proposed technique using the robot, Dry-AIREC, and with our learning model, it could perform cooking eggs with unknown ingredients. The robot could change the method of stirring and direction depending on the status of the egg, as in the beginning it stirs in the whole pot, then subsequently, after the egg started being heated, it starts flipping and splitting motion targeting specific areas, although we did not explicitly indicate them.

While deep learning enables real robots to perform complex tasks had been difficult to implement in the past, the challenge is the enormous amount of trial-and-error and motion teaching in a real environment. The manipulation of moving objects, due to their dynamic properties, requires learning a wide range of factors such as the object's position, movement speed, and grasping timing. We propose a data augmentation method for enabling a robot to grasp moving objects with different speeds and grasping timings at low cost. Specifically, the robot is taught to grasp an object moving at low speed using teleoperation, and multiple data with different speeds and grasping timings are generated by down-sampling and padding the robot sensor data in the time-series direction. By learning multiple sensor data in a time series, the robot can generate motions while adjusting the grasping timing for unlearned movement speeds and sudden speed changes. We have shown using a real robot that this data augmentation method facilitates learning the relationship between object position and velocity and enables the robot to perform robust grasping motions for unlearned positions and objects with dynamically changing positions and velocities.

We propose generalized equilibria of a three-dimensional color-gradient lattice Boltzmann model for two-component two-phase flows using higher-order Hermite polynomials. Although the resulting equilibrium distribution function, which includes a sixth-order term on the velocity, is computationally cumbersome, its equilibrium central moments (CMs) are velocity-independent and have a simplified form. Numerical experiments show that our approach, as in Wen et al. [Phys. Rev. E 100, 023301 (2019)] who consider terms up to third order, improves the Galilean invariance compared to that of the conventional approach. Dynamic problems can be solved with high accuracy at a density ratio of 10; however, the accuracy is still limited to a density ratio of $1\,000$. For lower density ratios, the generalized equilibria benefit from the CM-based multiple-relaxation-time model, especially at very high Reynolds numbers, significantly improving the numerical stability., Comment: 23 pages, 8 figures

Bridging the gap between motion models and reality is crucial by using limited data to deploy robots in the real world. Deep learning is expected to be generalized to diverse situations while reducing feature design costs through end-to-end learning for environmental recognition and motion generation. However, data collection for model training is costly, and time and human resources are essential for robot trial-and-error with physical contact. We propose "Deep Predictive Learning," a motion learning concept that predicts the robot's sensorimotor dynamics, assuming imperfections in the prediction model. The predictive coding theory inspires this concept to solve the above problems. It is based on the fundamental strategy of predicting the near-future sensorimotor states of robots and online minimization of the prediction error between the real world and the model. Based on the acquired sensor information, the robot can adjust its behavior in real time, thereby tolerating the difference between the learning experience and reality. Additionally, the robot was expected to perform a wide range of tasks by combining the motion dynamics embedded in the model. This paper describes the proposed concept, its implementation, and examples of its applications in real robots. The code and documents are available at: https://ogata-lab.github.io/eipl-docs

Americium-beryllium (AmBe), a well-known tagged neutron source, is commonly used for evaluating the neutron detection efficiency of detectors used in ultralow background particle physics experiments, such as reactor neutrino and diffuse supernova neutrino background experiments. In particular, AmBe sources are used to calibrate neutron tagging by selecting the 4438-keV $\gamma$-ray signal, which is simultaneously emitted with a neutron signal. Therefore, analyzing the neutron and $\gamma$-ray emission properties of AmBe sources is crucial. In this study, we used the theoretical shape of a neutron energy spectrum, which was divided into three parts, to develop models of the energy spectrum and verify the results using experimental data. We used an AmBe source to measure the energy spectra of simultaneously emitted neutrons and $\gamma$-rays and determine the emission ratio of the neutrons with and without $\gamma$-ray emission. The measured spectrum was consistent with that obtained from the simulated result, whereas the measured emission ratio was significantly different from the corresponding simulated result. Here, we also discuss the feasibility of determining the neutron emission rates from the spectra divided into three parts., Comment: 9 pages, 11 figures, 2 tables

Bismuth germanium oxide ($\rm Bi_{4} Ge_{3} O_{12}$, BGO) scintillation crystals are widely used as detectors in the fields of particle physics and astrophysics due to their high density, and thus higher efficiency for gamma-ray detection. Owing to their good chemical stability, they can be used in any environment. For rare-event searches, such as dark matter and coherent elastic neutrino-nucleus scattering, BGO crystals are essential to comprehend the response of nuclear recoil. In this study, we have analyzed the events of neutron elastic scattering with oxygen in BGO crystals. Then, we have measured the quenching factor for oxygen recoil energy in the BGO crystal as a function of recoil energy by using a monoenergetic neutron source., Comment: 14 pages, 11 figures, 1 table

NEWAGE is a direction-sensitive dark matter search experiment with a three-dimensional tracking detector based on a gaseous micro time projection chamber. A direction-sensitive dark matter search was carried out at Kamioka Observatory with a total live time of 318.0 days resulting in an exposure of 3.18 kg$\cdot$days. A new gamma-ray rejection and a head-tail determination analysis were implemented for this work. No significant non-isotropic signal from the directional analysis was found and a 90% confidence level upper limit on spin-dependent WIMP-proton cross section of 25.7 pb for WIMP mass of 150 GeV/c2 was derived. This upper limit is the most stringent in the direction-sensitive dark matter searches., Comment: submitted to PTEP

Sufficiently perceiving the environment is a critical factor in robot motion generation. Although the introduction of deep visual processing models have contributed in extending this ability, existing methods lack in the ability to actively modify what to perceive; humans perform internally during visual cognitive processes. This paper addresses the issue by proposing a novel robot motion generation model, inspired by a human cognitive structure. The model incorporates a state-driven active top-down visual attention module, which acquires attentions that can actively change targets based on task states. We term such attentions as role-based attentions, since the acquired attention directed to targets that shared a coherent role throughout the motion. The model was trained on a robot tool-use task, in which the role-based attentions perceived the robot grippers and tool as identical end-effectors, during object picking and object dragging motions respectively. This is analogous to a biological phenomenon called tool-body assimilation, in which one regards a handled tool as an extension of one's body. The results suggested an improvement of flexibility in model's visual perception, which sustained stable attention and motion even if it was provided with untrained tools or exposed to experimenter's distractions.

Circadian rhythms are biological rhythms of approximately 24 h that persist even under constant conditions without environmental daily cues. The molecular circadian clock machinery generates the physiological rhythms, which can be transmitted into the downstream output system. Owing to the stochastic nature of the biochemical reactions, the oscillation period of circadian rhythms exhibited by individual organisms or cells is not constant on a daily basis with variations as high as 10 % in terms of the coefficient of variation. Although the fluctuations in circadian rhythm is measured through a reporter such as bioluminescence or fluorescence, experimentally confirming whether the fluctuations found in the reporter system are the same as those in the clock itself is challenging. This study numerically and analytically investigated a coupled system of a circadian clock and its output system, and then compared the fluctuations in the oscillation period of the two systems. We found that the amount of fluctuation in the output system is smaller than that in the circadian clock when the degradation rate of the molecules responsible for the output system was at typical values. The results obtained imply that the output system improves the accuracy of the circadian rhythm without the need for any special denoising processes.

Deep robot vision models are widely used for recognizing objects from camera images, but shows poor performance when detecting objects at untrained positions. Although such problem can be alleviated by training with large datasets, the dataset collection cost cannot be ignored. Existing visual attention models tackled the problem by employing a data efficient structure which learns to extract task relevant image areas. However, since the models cannot modify attention targets after training, it is difficult to apply to dynamically changing tasks. This paper proposed a novel Key-Query-Value formulated visual attention model. This model is capable of switching attention targets by externally modifying the Query representations, namely top-down attention. The proposed model is experimented on a simulator and a real-world environment. The model was compared to existing end-to-end robot vision models in the simulator experiments, showing higher performance and data efficiency. In the real-world robot experiments, the model showed high precision along with its scalability and extendibility.

A gaseous time-projection chamber (TPC) with a reconstructable $z$ coordinate for nuclear recoil tracks has been developed for dark-matter searches and $\alpha$ particle imagings in a low radioactivity background. A TPC with a sheet-resistor field cage shows a potential to detect charge and photons produced by tracks if the sheet resistor has optical transmittance for visible light, and determine a drift length from a time difference between these signals. In this study, an optical transmittance of sheet resistors was measured to be $24.5\pm0.1_{\rm stat}\pm0.6_{\rm syst}\%$ for the $\rm {CF}_4$ gas scintillating light using an $\alpha$-particle source. The number of photoelectrons is observed to be $\rm\sim20~p.e.$ for 5.3~MeV$\alpha$ with the existence of a sheet resistor in the $\rm CF_4$ gas. Then, it is discussed how many photoelectrons to be observable by using multi-alkali-cathode phototubes and SiPMs for near-infrared light in order to detect lower energy tracks., Comment: 10 pages, 8 figures

We achieved contact-rich flexible object manipulation, which was difficult to control with vision alone. In the unzipping task we chose as a validation task, the gripper grasps the puller, which hides the bag state such as the direction and amount of deformation behind it, making it difficult to obtain information to perform the task by vision alone. Additionally, the flexible fabric bag state constantly changes during operation, so the robot needs to dynamically respond to the change. However, the appropriate robot behavior for all bag states is difficult to prepare in advance. To solve this problem, we developed a model that can perform contact-rich flexible object manipulation by real-time prediction of vision with tactility. We introduced a point-based attention mechanism for extracting image features, softmax transformation for predicting motions, and convolutional neural network for extracting tactile features. The results of experiments using a real robot arm revealed that our method can realize motions responding to the deformation of the bag while reducing the load on the zipper. Furthermore, using tactility improved the success rate from 56.7% to 93.3% compared with vision alone, demonstrating the effectiveness and high performance of our method.

Deep learning provides a powerful framework for automated acquisition of complex robotic motions. However, despite a certain degree of generalization, the need for vast amounts of training data depending on the work-object position is an obstacle to industrial applications. Therefore, a robot motion-generation model that can respond to a variety of work-object positions with a small amount of training data is necessary. In this paper, we propose a method robust to changes in object position by automatically extracting spatial attention points in the image for the robot task and generating motions on the basis of their positions. We demonstrate our method with an LBR iiwa 7R1400 robot arm on a picking task and a pick-and-place task at various positions in various situations. In each task, the spatial attention points are obtained for the work objects that are important to the task. Our method is robust to changes in object position. Further, it is robust to changes in background, lighting, and obstacles that are not important to the task because it only focuses on positions that are important to the task.