간행물
로봇학회논문지 KCI 등재 The Journal of Korea Robotics Society
- 발행기관 한국로봇학회
- 자료유형 학술지
- 간기 계간
- ISSN 1975-6291 (Print)2287-3961 (Online)
- 수록기간 2006 ~ 2020
- 주제분류 공학 > 제어계측공학 공학 분류의 다른 간행물
- 십진분류KDC 559DDC 629
권호리스트/논문검색
제13권 제1호 (통권 제47호) (2018년 2월) 10건
1.
2018.02
서비스 종료(열람 제한)
The image-to-image translation is one of the deep learning applications using image data. In this paper, we aim at improving the performance of object transfiguration which transforms a specific object in an image into another specific object. For object transfiguration, it is required to transform only the target object and maintain background images. In the existing results, however, it is observed that other parts in the image are also transformed. In this paper, we have focused on the structure of artificial neural networks that are frequently used in the existing methods and have improved the performance by adding constraints to the exiting structure. We also propose the advanced structure that combines the existing structures to maintain their advantages and complement their drawbacks. The effectiveness of the proposed methods are shown in experimental results.
2.
2018.02
서비스 종료(열람 제한)
This paper proposes a convolutional neural network model for distinguishing areas occupied by obstacles from a LiDAR image converted from a 3D point cloud. The channels of a LiDAR image used as input consist of the distances to 3D points, the reflectivities of 3D points, and the heights of 3D points from the ground. The proposed model uses a LiDAR image as an input and outputs a result of a segmented LiDAR image. The proposed model adopts refinement modules with skip connections to segment a LiDAR image. The refinement modules with skip connections in the proposed model make it possible to construct a complex structure with a small number of parameters than a convolutional neural network model with a linear structure. Using the proposed model, it is possible to distinguish areas in a LiDAR image occupied by obstacles such as vehicles, pedestrians, and bicyclists. The proposed model can be applied to recognize surrounding obstacles and to search for safe paths.
3.
2018.02
서비스 종료(열람 제한)
The talking head (TH) indicates an utterance face animation generated based on text and voice input. In this paper, we propose the generation method of TH with facial expression and intonation by speech input only. The problem of generating TH from speech can be regarded as a regression problem from the acoustic feature sequence to the facial code sequence which is a low dimensional vector representation that can efficiently encode and decode a face image. This regression was modeled by bidirectional RNN and trained by using SAVEE database of the front utterance face animation database as training data. The proposed method is able to generate TH with facial expression and intonation TH by using acoustic features such as MFCC, dynamic elements of MFCC, energy, and F0. According to the experiments, the configuration of the BLSTM layer of the first and second layers of bidirectional RNN was able to predict the face code best. For the evaluation, a questionnaire survey was conducted for 62 persons who watched TH animations, generated by the proposed method and the previous method. As a result, 77% of the respondents answered that the proposed method generated TH, which matches well with the speech.
4.
2018.02
서비스 종료(열람 제한)
Planetary global localization is necessary for long-range rover missions in which communication with command center operator is throttled due to the long distance. There has been number of researches that address this problem by exploiting and matching rover surroundings with global digital elevation maps (DEM). Using conventional methods for matching, however, is challenging due to artifacts in both DEM rendered images, and/or rover 2D images caused by DEM low resolution, rover image illumination variations and small terrain features. In this work, we use train CNN discriminator to match rover 2D image with DEM rendered images using conditional Generative Adversarial Network architecture (cGAN). We then use this discriminator to search an uncertainty bound given by visual odometry (VO) error bound to estimate rover optimal location and orientation. We demonstrate our network capability to learn to translate rover image into DEM simulated image and match them using Devon Island dataset. The experimental results show that our proposed approach achieves ~74% mean average precision.
5.
2018.02
서비스 종료(열람 제한)
Recently, the safety in vehicle also has become a hot topic as self-driving car is developed. In passive safety systems such as airbags and seat belts, the system is being changed into an active system that actively grasps the status and behavior of the passengers including the driver to mitigate the risk. Furthermore, it is expected that it will be possible to provide customized services such as seat deformation, air conditioning operation and D.W.D (Distraction While Driving) warning suitable for the passenger by using occupant information. In this paper, we propose robust vehicle occupant detection algorithm based on RGB-Depth-Thermal camera for obtaining the passengers information. The RGB-Depth-Thermal camera sensor system was configured to be robust against various environment. Also, one of the deep learning algorithms, OpenPose, was used for occupant detection. This algorithm is advantageous not only for RGB image but also for thermal image even using existing learned model. The algorithm will be supplemented to acquire high level information such as passenger attitude detection and face recognition mentioned in the introduction and provide customized active convenience service.
6.
2018.02
서비스 종료(열람 제한)
In this paper, we propose Intelligent Driver Assistance System (I-DAS) for driver safety. The proposed system recognizes safety and danger status by analyzing blind spots that the driver cannot see because of a large angle of head movement from the front. Most studies use image pre-processing such as face detection for collecting information about the driver's head movement. This not only increases the computational complexity of the system, but also decreases the accuracy of the recognition because the image processing system dose not use the entire image of the driver's upper body while seated on the driver's seat and when the head moves at a large angle from the front. The proposed system uses a convolutional neural network to replace the face detection system and uses the entire image of the driver's upper body. Therefore, high accuracy can be maintained even when the driver performs head movement at a large angle from the frontal gaze position without image pre-processing. Experimental result shows that the proposed system can accurately recognize the dangerous conditions in the blind zone during operation and performs with 95% accuracy of recognition for five drivers.
7.
2018.02
서비스 종료(열람 제한)
This paper presents a vision-based fall detection system to automatically monitor and detect people’s fall accidents, particularly those of elderly people or patients. For video analysis, the system should be able to extract both spatial and temporal features so that the model captures appearance and motion information simultaneously. Our approach is based on 3-dimensional convolutional neural networks, which can learn spatiotemporal features. In addition, we adopts a thermal camera in order to handle several issues regarding usability, day and night surveillance and privacy concerns. We design a pan-tilt camera with two actuators to extend the range of view. Performance is evaluated on our thermal dataset: TCL Fall Detection Dataset. The proposed model achieves 90.2% average clip accuracy which is better than other approaches.
8.
2018.02
서비스 종료(열람 제한)
In this paper, a method for estimation of external force on an end-effector using joint torque sensor is proposed. The method is based on portion of measure torque caused by external force. Due to noise in the torque measurement data from the torque sensor, a recursive least-square estimation algorithm is used to ensure a smoother estimation of the external force data. However it is inevitable to create a delay for the sensor to detect the external force. In order to reduce the delay, modified recursive least-square is proposed. The performance of the proposed estimation method is evaluated in an experiment on a developed six-degree-of-freedom robot. By using NI DAQ device and Labview, the robot control, data acquisition and The experimental results output are processed in real time. By using proposed modified RLS, the delay to estimate the external force with the RLS is reduced by 54.9%. As an experimental result, the difference of the actual external force and the estimated external force is 4.11% with an included angle of 5.04° while in dynamic state. This result shows that this method allows joint torque sensors to be used instead of commonly used external sensory system such as F/T sensors.
9.
2018.02
서비스 종료(열람 제한)
This paper describes a study on posture control of the multi-legged biomimetic underwater robot (CALEB10). Because the underwater environment has a feature that all degrees of freedom are coupled to each other, we designed the posture control algorithm by separating each degree of freedom. Not only should the research on posture control of underwater robots be a precedent study for position control, but it is also necessary to compensate disturbance in each direction. In the research on the yaw directional posture control, we made the drag force generated by the stroke of the left leg and the right leg occur asymmetrically, in order that a rotational moment is generated along the yaw direction. In the composite swimming controller in which the controllers in each direction are combined, we designed the algorithm to determine the control weights in each direction according to the error angle along the yaw direction. The performance of the proposed posture control method is verified by a dynamical simulator and underwater experiments.
10.
2018.02
서비스 종료(열람 제한)
This paper proposes a Dual Controller System for Precision Control (DCSPC) for control of the gripper. The DCSPC consists of two subsystems, CDSP (Controller based DSP) and CARM (Controller based ARM processor). The CDSP is developed on a DSP processor and controls the gripping motor and LVDT. In particular, the CARM is implemented using Linux and ARM processor according to recent research related to open-source. The robot for high-precision assembly is divided into the robot control and the gripper control section and controls CARM and CDSP systems respectively. In this paper, we also proposed and measured the performance of communication API. As a result, it is expected to recognize improvements in communication between CARM and the robot controller, and will continue to conduct relevant research among other commercial robot controllers.
