Search

Probabilistic localization based on Bayesian theory has been researched as a sensor fusion method to improve the robustness of localization. Pieces of position information, generated by sensors’ observation models with consideration for noises, are fused according to Bayesian theory. However, having large noises not considered in their observation models, the sensors output erroneous position information; thus, the fusion result has a significant error, even when the other sensors output correct ones. In this research, we have proposed a sensor fusion system with a relative correlation checking test to realize robust localization. Pieces of erroneous position information, biased against others and having a negative correlation with others, are detected and excluded in our proposed system by checking their correlation between all of them. The purpose of this paper is to evaluate the robustness of our fusion system by conducting recursive localization experiments in various environments.

In this paper, we propose a new system for operator support in teleoperation using Augmented Reality(AR) based on sensing data . In the proposed system, an 360-degree camera is equipped with a mobile machine in a remote place which is controlled by an operator, and the captured 360-degree image is displayed to the head mounted display put on the head of the operator. The operator can control the machine smoothly by watching what he needs for his current task by turning his head. In addition, the provided image around the machine has no blind spot because of the 360-degree image. However, a large transfer delay or a dropped frame occurs due to the enormous data size of the 360-degree image, which adversely influences the operator performance. In order to solve this problem, the AR image is actually provided to the operator, in which CG objects generated based on other sensor information are superimposed on the 360-degree image as AR objects. In this paper, the usefulness of the proposed method is shown through the experimental results on obstacle avoidance of a mobile robot by teleoperation.

In this paper, we propose a new teleoperation system of a mobile robot using Augmented Reality(AR) based on 3D LiDAR information for operator support. In the proposed system, an 360-degree camera is equipped with a robot which works in a remote place controlled by an operator, and the captured 360-degree image is displayed to the head mounted display put on the head of the operator. The operator can smoothly control the robot watching what he needs for his task by turning his head. In addition, the provided image around the robot has no blind spot because of the 360-degree image. However, the data size of the 360-degree image is enormous and it causes a large transfer delay or dropped frames which adversely influence the operator performance. In order to solve this problem, it is proposed that an AR image is actually provided to the operator, in which objects generated based on 3D LiDAR information are superimposed on the 360-degree image as AR objects. In this paper, the usefulness of the proposed method is shown through the experimental results on parking the robot remotely.

With increase in the number of elderly people in the Japanese society, traffic accidents caused by elderly driver is considered problematic. The primary factor of the traffic accidents is a reduction in their driving cognitive performance. Therefore, a system that supports the cognitive performance of drivers can greatly contribute in preventing accidents. Recently, the development of devices for visually providing information, such as smart glasses or head up display, is in progress. These devices can provide more effective supporting information for cognitive performance. In this article, we focus on the selection problem of information to be presented for drivers to realize the cognitive support system. It has been reported that the presentation of excessive information to a driver reduces the judgment ability of the driver and makes the information less trustworthy. Thus, indiscriminate presentation of information in the vision of the driver is not an effective cognitive support. Therefore, a mechanism for determining the information to be presented to the driver based on the current driving situation is required. In this study, the object that contributes to execution of avoidance driving operation is regarded as the object that drivers must recognize and present for drivers. This object is called as contributing object. In this article, we propose a method that selects contributing objects among the appeared objects on the current driving scene. The proposed method expresses the relation between the time series change of an appeared object and avoidance operation of the driver by a mathematical model. This model can predict execution timing of avoidance driving operation and estimate contributing object based on the prediction result of driving operation. This model named as contributing model consisted of multi-hidden Markov models. Hidden Markov model is time series probabilistic model with high readability. This is because that model parameters express the probabilistic distribution and its statistics. Therefore, the characteristics of contributing model are that it enables the designer to understand the basis for the output decision. In this article, we evaluated detection accuracy of contributing object based on the proposed method, and readability of contributing model through several experiments. According to the results of these experiments, high detection accuracy of contributing object was confirmed. Moreover, it was confirmed that the basis of detected contributing object judgment can be understood from contributing model.

The authors aim to realize a wearable robot that supports the motion of entire torso. The first task to realize the robot is to determine its structure suitable for motion support. Focusing on the body surface where the robot makes direct contact, we examine the robot structure that can follow the deformation of the body surface. In order to clarify the robot structure, it is necessary to measure the deformation of the body surface. However, with the existing measurement method of body surface deformation using markers, the robot structure can not be clarified. Hence, we have studied a new system to measure the deformation of the body surface without using any markers. In this system, the body surface shape is obtained as time-series point cloud data. After that, the body surface (point cloud data) is divided into multiple local areas, and the time-series correspondence of each local area is clarified. Finally, the deformation is measured by tracking each local area. In this paper, the details of the system are described, and the measurement accuracy of the system is discussed. Furthermore, we quantitatively clarify how the body surface of the torso deforms at the time of forward bending by using the system.

This paper evaluates our measuring position determination method of Unmanned Aerial Vehicles (UAVs) for automatic visual inspection through an experimentation in real world. This method is under the assumption that the 3D model of a target facility, required accuracy of gathering data and specs of instruments are given. The triangular polygons for shape description of the 3D model are reconstructed for measurement and the measuring positions: the waypoints that an UAV travels to gather the accurate data of the target facility efficiently are calculated systematically. For basic evaluation of this method, we conducted an experimentation in real world. The measuring positions were calculated for an actual target and a camera. An experimenter brought a camera to the measuring positions and gathered image data. This experimentation result showed the effectivity of the method.

This paper discusses a robust localization method that uses particle filtering. A particle filter can suppress the influence of temporary noise on a sensor based on past sensor data. However, localization fails when a sensor is affected by noise that lasts for several minutes even when using a particle filter. We have previously proposed a majority rule-based sensor fusion system that removes sensors affected by such constant noise. To date, we have conducted some experiments to evaluate the effectiveness of this system without particle filter. In this paper, we experimentally confirm the effectiveness of this system with particle filtering. The results show that the proposed system reduces position errors in part of the environment.

This paper presents a measuring position determination method of Unmanned Aerial Vehicles (UAVs) for automatic visual inspection. Ensuring accuracy and considering efficiency of operations in data gathering to detect anomaly are both important functions for inspections. This method is under the assumption that the 3D model of a target facility, required accuracy of gathering data and specs of instruments are given. The triangular polygons for shape description of the 3D model are reconstructed to those for measurement and the measuring positions; the waypoints that an UAV travels to gather the accurate data of the target facility efficiently are calculated systematically. If the measuring positions are obtained, the flight path appropriating for visual inspection can be planned by general path-planning methods by setting the measuring positions as waypoints. For evaluation of efficiency of measurement, the proposed method is examined through computer simulations on the assumption that an UAV measure image data of two typical bridges; a girder bridge and an arch bridge.

Robot motions are designed according to mechanical structures such as frame structure, number of joints and so on. Therefore, structure changes such as joint failures or frame distortion make the designed motions unaccomplished. In service robots working for general users, it is desired to recover the motions autonomously. In trajectory controls, it is desired to realize motions with some trajectories even on the changed robot structures. Conventionally, motion adaptation methods with inverse kinematics and identification of changed structures have been used to solve the problems. But, for multi-degree-of-freedom robots such as service robots, the methods are difficult to apply. New motion adaptation methods are proposed in this paper. In the methods, all motions are expressed as time series of joint angles, and corresponding joint angles are generated efficiently on changed structure based on exploration methods. Representing postures are sorted by the Vector Quantization from the series of designed joint angles. Corresponding joint angles are found by the Simulated Annealing only on the representing postures because of reduction of exploration costs. Adapted motions to realize the intended trajectories on the changed structures are estimated by the Kriging methods. In simulation, the adapted motions which achieve the intended trajectories are generated autonomously without identification. Its exploration costs are reduced about 1/7 comparing with full search method.

We suggest motion adaptation method against environmental changes for service robots. Several motions such as walking, dancing, demonstration and so on are described with time series patterns. These motions are optimized with the architecture of the robot and under certain surrounding environment. Under unknown environment, robots cannot accomplish their tasks. We propose motion generation system based on heuristic search with histories of internal sensor values. New motion patterns are explored under unknown environment based on self-evaluation function. Self-evaluation function is composed of difference of internal sensor values between standard environment and unknown environment. New motion patterns are generated to maximize self-evaluation function without external information, such as run length, position of robot, human observation and so on. Experimental results show that the possibility to adapt autonomously patterned motions to environmental changes.

Robot motions are designed according to the properties such as the number of joints, the range of each joint, the length of each frame and so on. Structure changes such as joint failures or frame distortion make the tasks of prepared motions unaccomplished. Identification of structure changes on multi degree-of-freedom (DOF) robots like humanoid is so difficult that the traditional approaches based on inverse kinematics would not be applied. We propose a motion adaptation method to generate new proper motions accomplishing the tasks on changed structure without identification of the structure changes. All the motions can be expressed with the time-series data of postures. Each proper posture as an element of motions for changed structure is estimated by Kriging with joint angle data of several original postures sorted by Vector Quantization (VQ) and the corresponding postures explored on changed structure by Simulated Annealing (SA). At the case the tasks couldn't be accomplished with the generated motions in designed DOF, new motions are generated in expanded DOF adding redundancy joints. Experimental results show that the generated motion by proposed method is accomplished 96.6% similarity with the original task referenced failure motion's task to 0% on changed structure by single joint locking.