1P1-A8 親子型群ロボットシステムにおける物体探索および協調搬送(81. スーパーメカノシステムII)

Handling the deformations of the lung tissues in successive chest computed tomography (CT) scans of a patient is a vital step in any computer-aided diagnostic (CAD) system for lung cancer diagnosis. In this paper, we propose a new nonrigid registration methodology for the segmented lung region from CT data that involves two steps. The first step globally aligns the target-to-reference CT scans using an affine transformation based on ascent maximization of the estimated mutual information of the calculated distance map using the fast marching level sets method inside the segmented lung for both the target and reference objects. The second step is the local alignment of the target lung object in order to correct for intricate relative deformations due to breathing and heart beats. The local deformations are handled based on displacing each voxel of the target over evolving closed equi-spaced surfaces (iso-surfaces) to closely match the reference object. In order to displace the voxel on the iso-surfaces of the target lung object, the initial voxel-to-voxel match between target and reference lung objects is estimated by solving the 3D Laplace equation between each two corresponding iso-surfaces on the reference and target objects. Finally, the estimation of voxel-to-voxel match is refined through iterative energy minimization using a generalized Gauss-Markov random field (GGMRF) model. Qualitative and quantitative results demonstrate the promise of the proposed nonrigid registration framework.

The existence of handle affordances has been classically demonstrated using the Stimulus-Response Compatibility paradigm, with shorter response times when the orientation of the object handle and the response hand are compatible in comparison to incompatible. Yet the activation of handle affordances from visual objects has been investigated in very simple situations involving single stimulus and motor response. As natural perceptual scenes are usually composed of multiple objects that could activate multiple affordances, the consequence of multiple affordance activation on the perception and processing of a given object of the scene requires more investigation. The aim of this study was to determine the impact of distractor affordances on the processing of a target object in situations involving several familiar graspable objects. In two online experiments, 229 participants had to select a target object (the kitchen utensil or the tool) in a visual scene displaying a pair of objects. They performed left key presses when the target was on the left and right key presses when the target was on the right. Target handle orientation and response side could be compatible or incompatible. Critically, target and distractor objects had similar or dissimilar handle affordances, with handles oriented for left- or right-hand grasps. Results from the two experiments showed slower response times when target and distractor objects had similar handle affordances in comparison to dissimilar affordances, when participants performed right hand responses and when target orientation and response were compatible. Thus, affordance similarity between objects may interfere rather than facilitate object processing and slow down target selection. These findings are in line with models of affordance and object selection assuming automatic inhibition of distractors' affordances for appropriate object interaction.

This paper describes how two cameras were used to implement stereoscopic visual imaging through a robot vision system. Image processing was applied to construct a real-time stereoscopic image tracking system for use in robots. The target object was sought by image preprocessing. The deviation of the target object from the optical center of the two cameras was measured to calculate the stereoscopic image depth of the target object, and the target object was instantly locked for real-time tracking. This system consists of two parts. The first part uses depth calculation to determine the distance between the target object and robot platform and applies image processing to simplify the image. The deviation angles of the left and right images are subsequently measured to calculate the target object distance. The second part uses real-time image tracking so that the platform can lock the image of the target object into an overlapping image. Proportional–integral–derivative (PID) feedback controls the robot movement and instantly obtains the target object and offset direction. The signal is exported to dc servo motors to drive the rotation of the two cameras and the movement of the platform. The image of the target object is locked in the image center of the stereoscopic visual system for real-time tracking.

In 4 picture-word interference experiments, speakers named a target object that was presented with a context object. Using auditory distractors that were phonologically related or unrelated either to the target object or the context object, the authors assessed whether phonological processing was confined to the target object or not. Phonological activation of the context objects was reliably observed if the target and context objects were embedded in a conceptually coherent scene (e.g., if the picture showed a mouse eating some cheese), regardless of whether the target was cued by its thematic role (agent vs. patient) or by color. However, this activation dissipated if the two objects were presented in an arbitrary object array (e.g., if the cheese was presented along with a finger). These findings suggest that conceptual coherence among multiple objects affects the information flow in the conceptual-lexical system during speech planning.

Previous studies in our laboratory examining pointing and reach-to-grasp movements of Parkinson's disease patients (PDPs) have found that PDPs exhibit specific deficits in movement coordination and in the sensorimotor transformations required to accurately guide movements. We have identified a particular difficulty in matching unseen limb position, sensed by proprioception, with a visible target. In the present work, we further explored aspects of complex sensorimotor transformation and motor coordination using a reach-to-grasp task in which object shape, visual feedback, and dopaminergic medication were varied. Normal performance in this task requires coordinated generation of appropriate reach, to bring the hand to the target, and differentiated grasp, to preshape the hand congruent with object form. In Experiment 1, we tested PDPs in the off-medication state. To examine the dependence of subjects on visual feedback and their ability to implement intermodal sensory integration, we required them to reach and grasp the target objects in three conditions: (1) Full Vision, (2) Object Vision with only the target object visible and, (3) No Vision with neither the moving arm nor the target object visible. PDPs exhibited two types of deficits. First, in all conditions, they demonstrated a generalized slowing of movement or bradykinesia. We consider this an intensive deficit, since it involves largely a modulation of the gain of specific task parameters: in this case, velocity of movement. Second, they were less able than controls to extract critical proprioceptive information and integrate it with vision in order to coordinate the reach and grasp components of movement. These deficits which involve the coordination of different inputs and motor components, we classify as coordinative deficits. As in our previous work, the PDPs' deficits were most marked when they were required to use proprioception to guide their hand to a visible target (Object Vision condition). But even in the full-vision condition, their performance only became fully accurate when both the target and effector (hand) were simultaneously visible. In Experiment 2, PDPs were tested on their dopaminergic replacement therapy. Dopaminergic treatment significantly ameliorated the bradykinesia of the PDPs, but produced no changes in the hand preshaping deficiencies of PDPs. These results suggest that adequate treatment of the PDPs may more readily compensate for intensive, than coordinative deficits, since the latter are likely to depend on specific and time-dependent neural interdependencies that are unlikely to be remediated simply by increasing the gain of a pathway.

MSST-ResNet: Deep multi-scale spatiotemporal features for robust visual object tracking

Recent adaptive memory and evolutionary metaheuristics for mixed integer programming have included proposals for introducing inequalities and target objectives to guide the search. These guidance approaches are useful in intensification and diversification strategies related to fixing subsets of variables at particular values, and in strategies that use linear programming to generate trial solutions whose variables are induced to receive integer values. In Part I (the present paper), we show how to improve such approaches by new inequalities that dominate those previously proposed and by associated target objectives that underlie the creation of both inequalities and trial solutions. Part I focuses on exploiting inequalities in target solution strategies by including partial vectors and more general target objectives. We also propose procedures for generating target objectives and solutions by exploiting proximity in original space or projected space. Part II of this study (to appear in a subsequent issue) focuses on supplementary linear programming models that exploit the new inequalities for intensification and diversification, and introduce additional inequalities from sets of elite solutions that enlarge the scope of these models. Part II indicates more advanced approaches for generating the target objective based on exploiting the mutually reinforcing notions of reaction and resistance. Our work in the concluding segment, building on the foundation laid in Part I, examines ways our framework can be exploited in generating target objectives, employing both older adaptive memory ideas of tabu search and newer ones proposed here for the first time.

Underwater target acquisition and identification performed by manipulators having broad application prospects and value in the field of marine development. Conventional manipulators are too heavy to be used for small target objects and unsuitable for shallow sea working. In this paper, a bio-inspired Father–Son Underwater Robot System (FURS) is designed for underwater target object image acquisition and identification. Our spherical underwater robot (SUR), as the father underwater robot of the FURS, has the ability of strong dynamic balance and good maneuverability, can realize approach the target area quickly, and then cruise and surround the target object. A coiling mechanism was installed on SUR for the recycling and release of the son underwater robot. A Salamandra-inspired son underwater robot is used as the manipulator of the FURS, which is connected to the spherical underwater robot by a tether. The son underwater robot has multiple degrees of freedom and realizes both swimming and walking movement modes. The son underwater robot can move to underwater target objects. The vision system is installed to enable the FURS to acquire the image information of the target object with the aid of the camera, and also to identify the target object. Finally, verification experiments are conducted in an indoor water tank and outdoor swimming pool conditions to verify the effectiveness of the proposed in this paper.

In the humanoid robot system, we used two computers responsible for visual information processing system and motion control system respectively. The visual platform of the humanoid robot is designed the same as the human head, which can search for the target object in a wide range. The target object tracking process consist of four cycle steps: search, match, refreshing and forecasting; the grasping object process contains five steps: find and coarse positioning target object, walking close to the target object, based on the visual feedforward coarse alignment, fine alignment based on visual feedback, grasp. Each step requires a different control strategy. In experiments, we use the motion capture system to measure the three-dimensional position of the target object, and as a benchmark to measure the measurement accuracy of the humanoid robot vision platform. Experiments show that based on the visual platform, the visual feedforward and the visual feedback control strategy, the target object tracking accuracy has been improved.

Method for detecting subject area target objects is proposed, which uses machine learning algorithms for adaptive object recognition, taking into account the specifics of the subject area, which allows to significantly reduce the time of data processing and reduce the risk of losing important information. The method for detecting subject area target objects allows to transform input data in the form of the researched text and a pre-processed and balanced corpus of texts of the researched subject area into output data in the form of a formed set of target objects from the researched text, which is a combined set of keywords found by various methods without repetitions and the set of NER grouped by lemmatization. The proposed method for detecting subject area target objects differs from the existing ones by taking into account keywords and noun entities of the subject area, which made it possible to increase accuracy of detection of detecting subject area target objects as result of taking into account noun entities. To research the effectiveness of developed method for detecting subject area target objects, a training dataset of 400 texts in the Ukrainian language was formed. Also, for the validation of proposed method, the software was developed to convert the text content of files from the test sample into a set of target objects of the subject area; separate console software was created to use the obtained list of target objects for the studied texts and dictionaries from the subject areas selected according to the dataset. The performed study of the effectiveness of the developed method for detecting subject area target objects revealed that the target objects of the subject areas found by the method are able to perform the further task of classification, demonstrating on the metric of Euclidean distances the grouping of texts of the same category and the increase of the distance orthogonal to it. This determines the beneficial effect and scope of the developed method.

Weakly electric fish use self-generated electric fields to identify prey, explore their surroundings, and communicate with conspecifics. They have electroreceptors over their entire skin surfaces, and readings from these electroreceptors form electric images. The characteristics of electric images include important information such as the locations, shapes, and electrical properties of target objects. The lateral distance to a target object can be estimated using the rate between maximal slope and maximal amplitude, or relative slope, and provides a direct cue to localize target objects, irrespective of size and electrical properties. Sensor readings acquired from distributions of electroreceptors are interpreted as spatial electric images. Temporal electric images can be assessed using a single electroreceptor by back-and-forth swimming. Some species of weakly electric fish engage in back-and-forth swimming behavior during foraging and can use spatiotemporal information to localize objects. With active body movement and forward swimming, a single electroreceptor can scan the time course of potential perturbation caused by a target object. If there is more than one target object, electric images may be distorted by interactions between individual images. In this paper, we suggest that temporal relative slope is an effective measure for the electrolocation of multiple objects by robotic systems. We investigate the characteristics of temporal relative slopes based on back-and-forth sweeping motions to localize multiple spherical objects via modeling experiments. This method can be applied to biomimetic robotic systems in the underwater environment.

In peer-to-peer (P2P) overlay networks, objects are distributed to peers in various ways like downloading. Service supported by an object is modeled to be a set of methods and quality of service (QoS). In addition to discovering a peer holding a target object, it is critical to discuss what peer is allowed to manipulate the target object in what method. In this paper, we take an acquaintance approach to finding and manipulating objects in P2P overlay networks. An acquaintance peer of a peer p is a peer whose service the peer p knows and with which the peer p can directly communicate. If an access request is issued, a peer has to find where a target object exists, how to obtain an access right on the target object, and how to manipulate the target object. In this paper, we discuss ways to obtain results satisfying an access request by the cooperation of acquaintances. Acquaintance peers of a peer p may hold different information on target peers since it takes time to propagate change information of the target peers and peers may be faulty. Here, it is critical to discuss how much a peer can trust each acquaintance. First, we define how much a requesting peer is satisfied for each access request issued to an acquaintance. Then, we define the trustworthiness of an acquaintance peer based on the satisfiability. That is, a peer considers an acquaintance to be more trustworthy if the acquaintance returns more satisfiable replies.

The Universal Jamming Gripper, made up of a coffee grain-filled balloon membrane and a vacuum pump, operates by applying an unvarying or constant negative pressure set by the manipulator on the target object for gripping, not taking into account the suitable amount of negative pressure depending on the size and weight of the target object. With the constant or same amount of negative pressure applied on the target objects of different sizes and weight, the gripper can diminish the structural integrity of some of these target objects. To eradicate this problem, the researchers developed a system to vary the negative pressure of the Universal Jamming Gripper through an artificial neural network depending on the size of the object, obtained through a vision-based object classification scheme, and weight obtained from a load cell connected to the computer. An Artificial Neural Network (ANN), trained with three inputs, such as the pixel area of one side of the target object, the pixel area of another side of the object and its weight, is used to automatically determine the optimum negative pressure needed to successfully grip the target object. After testing and experimentation, the ANN is proven to output the optimum negative pressure needed to successfully conform to and grip the target object, as evident with the 99.131% result from testing, based on the regression plot from MatLab.

In a cooperation task by multiple robots, it may happen that a working robot can not detect a target object for handling due to a sensor occlusion. In this situation, if another cooperative robot observes the working robot with the target object and detects their positions and orientations, it will be possible for the working robot to complete the handling task. Such behavior is a kind of indirect cooperation. This study proposes a method for such an indirect cooperation based on an observation by the partner robot. The observing robot obtains corresponding points of Scale-Invariant Feature Transformation (SIFT) on the working robot with hand and the target object from multiple captured images. The 3-D position of the target object and hand motion of the working robot can be detected by applying stereo vision theory to the points. The working robot is then able to get the relation between its hand and the target object indirectly from the observing robot. We describe each process to establish the indirect cooperation. Fundamental experiments confirmed the validity of presented method.

In this paper, a method using a stereo vision device and fuzzy control to guide a robot arm to grasp a target object is proposed. The robot arm has five degrees of freedom including a gripper and four joints. The stereo vision device located beside the arm captures images of the target and the gripper. Image processing techniques such as color space transformation, morphologic operation, and 3-D position measurement are used to identify the target object and the gripper from the captured images and estimate their relative positions. Based on the estimated positions of the gripper and the target, the gripper can approach and grasp the target using inverse kinematics. However, since the robot arm’s accuracy of movement may be affected by gearbox backlash or hardware uncertainty, the gripper might not approach the desired position with precision using only inverse kinematics. Therefore, a fuzzy compensation method is added to correct any position errors between the gripper and target such that the gripper can grasp the target. Using the proposed method, the stereo vision device can not only locate the target object but also trace the position of the robot arm until the target object is grasped. Finally, some experiments are conducted to demonstrate successful implementation of the proposed method on the robot arm control.