• The Journal of Korea Robotics Society
• /
• v.17 no.4
• /
• pp.425-430
• /
• 2022

As robots are considered one of the mainstream digital transformations, robots with machine vision becomes a main area of study providing the ability to check what robots watch and make decisions based on it. However, it is difficult to find a small object in the image mainly due to the flaw of the most of visual recognition networks. Because visual recognition networks are mostly convolution neural network which usually consider local features. So, we make a model considering not only local feature, but also global feature. In this paper, we propose a detection method of a small marker on the object using deep learning and an algorithm that considers global features by combining Transformer's self-attention technique with a convolutional neural network. We suggest a self-attention model with new definition of Query, Key and Value for model to learn global feature and simplified equation by getting rid of position vector and classification token which cause the model to be heavy and slow. Finally, we show that our model achieves higher mAP than state of the art model YOLOr.

• Journal of the Korean Geotechnical Society
• /
• v.21 no.2
• /
• pp.121-126
• /
• 2005

It is not uncommon to observe tilt of buildings and towers as a result of unexpected differential foundation settlements. Over the years, a number of engineering methods including the soil extraction technique have been attempted to reduce inclination of buildings and towers. In this research, a series of novel geotechnical centrifuge model tests by using a state-of-the-art in-flit robotic manipulator have been conducted to study key factors which govern the restoration of building tilts. In the centrifuge model tests, the robotic manipulator was used to drill and extract soil in-flight near an initially tilted model building. The soil extraction was to induce stress release, thereby mitigating the inclination of the model building. Insights into the effects of different configurations, soil density and sequences of drilling observed during the centrifuge model tests on the restoration of the model building are to be investigated.

• Cartoon and Animation Studies
• /
• s.49
• /
• pp.537-561
• /
• 2017

Machines and robotic structures are questioning existential bases of human beings. They influence our way of thinking and transform our social philosophies and value systems. The same time they keep their fascination ever since. Mechanical technique and skills are symbols for development and hazard at the same time. The attraction of this ambivalence is reviewed in this thesis by having a look at kinetic elements in modern sculpture. The author is focusing on classical sculptural positions that use machines or machinery within sculptural artworks. The historical development is examined and certain examples from the classical modern era are discussed as role models during their time. On this scenario, the portraits of two young contemporary outstanding positions in the field of kinetic art and installation are analyzed and explained. One of the goals was to investigate the eastern and the western language of two artists that are doing artworks in the same field and compare the works and the cultural expressions to each other. Different cultural positions and visual languages become visible due to the research. And the author's final conclusions is, that even in a globalized world there will always be significant local distinguishes remaining.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.335-349
• /
• 2023

Bridges constantly undergo deterioration and damage, the most common ones being concrete damage and exposed rebar. Periodic inspection of bridges to identify damages can aid in their quick remediation. Likewise, identifying components can provide context for damage assessment and help gauge a bridge's state of interaction with its surroundings. Current inspection techniques rely on manual site visits, which can be time-consuming and costly. More recently, robotic inspection assisted by autonomous data analytics based on Computer Vision (CV) and Artificial Intelligence (AI) has been viewed as a suitable alternative to manual inspection because of its efficiency and accuracy. To aid research in this avenue, this study performs a comparative assessment of different architectures, loss functions, and ensembling strategies for the autonomous segmentation of bridge components and damages. The experiments lead to several interesting discoveries. Nested Reg-UNet architecture is found to outperform five other state-of-the-art architectures in both damage and component segmentation tasks. The architecture is built by combining a Nested UNet style dense configuration with a pretrained RegNet encoder. In terms of the mean Intersection over Union (mIoU) metric, the Nested Reg-UNet architecture provides an improvement of 2.86% on the damage segmentation task and 1.66% on the component segmentation task compared to the state-of-the-art UNet architecture. Furthermore, it is demonstrated that incorporating the Lovasz-Softmax loss function to counter class imbalance can boost performance by 3.44% in the component segmentation task over the most employed alternative, weighted Cross Entropy (wCE). Finally, weighted softmax ensembling is found to be quite effective when used synchronously with the Nested Reg-UNet architecture by providing mIoU improvement of 0.74% in the component segmentation task and 1.14% in the damage segmentation task over a single-architecture baseline. Overall, the best mIoU of 92.50% for the component segmentation task and 84.19% for the damage segmentation task validate the feasibility of these techniques for autonomous bridge component and damage segmentation using RGB images.

• Biomaterials and Biomechanics in Bioengineering
• /
• v.2 no.4
• /
• pp.249-269
• /
• 2015

A problem faced by oil companies is the maintenance of the location register of pipelines that cross the surf zone, the regular survey of their location, and also their inspection. A survey of the state of art did not allow identifying operating systems capable of executing such tasks. Commercial technologies available on the market also do not address this problem and/or do not satisfy the presented requirements. A possible solution is to use robotic systems which have the ability to walk on the shore and in the surf zone, subject to existing currents and ripples, and being able to withstand these ambient conditions. In this sense, the authors propose the development of a spider crab biologically inspired robot to achieve those tasks. Based on these ideas, this work presents a biomechanical study of the spider crab, its modeling and simulation using the SimMechanics toolbox of Matlab/Simulink, which is the first phase of this more vast project. Results show a robot model that is moving in an "animal like" manner, the locomotion, the algorithm presented in this paper allows the crab to walk sideways, in the desired direction.

• Korean Journal of Agricultural Science
• /
• v.42 no.3
• /
• pp.269-276
• /
• 2015

Unmanned Aerial Vehicle (UAV) has several advantages over conventional remote sensing techniques. They can acquire high-resolution images quickly and repeatedly. And with a comparatively lower flight altitude, they can obtain good quality images even in cloudy weather. In this paper, we discussed the state-of-the-art of the domestic and international use of UAV in agricultural sector as well as assessed its utilization and applicability for agricultural environment in Korea. Association of robotic, computer vision and geomatic technologies have established a new paradigm of low-altitude aerial remote sensing that has now been receiving attention from researchers all over the world. In a field study, it has been found that use of UAV imagery in an agricultural subsidy program can reduce the farmers' complain and provide objective evidence. UAV high resolution photography can also be helpful in monitoring the disposal zone for animal carcasses. Due to its expeditiousness and accuracy, UAV imagery can be a very useful tool to evaluate the damage in case of an agricultural disaster for both parties insurance companies and the farmers. Also high spatial and temporal resolution in UAV system can increase the prediction accuracy which in turn help to maintain the agricultural supply and demand chain.

• International Journal of High-Rise Buildings
• /
• v.9 no.1
• /
• pp.95-104
• /
• 2020

Recently, robotic technologies have significantly improved, bringing considerable enhancements in many sectors; the main objective of this paper is to figure out if these innovations have also involved the building industry. To achieve this purpose, it has been considered crucial to first reshape and clarify some concepts, incorporating a much more flexible understanding of the term "robot", as well as the formulation of its future potential. Subsequently, it has been carried out an analysis of the various advanced devices that are currently available to be employed in the construction processes; the review includes a thorough classification of construction robots, divided into 18 families reflecting their purpose of use, and a dissection based on the term used to define them. The attention has been focused on the most updated and recent robots and, in their absence, on the most advanced machines prevailing. This operation has been achieved taking into account the development history of construction robots, as well as the analyses and classifications previously conducted, reconsidering them according to the just mentioned reflections. Furthermore, an in-depth exploration of the exoskeletons, as well as on a sophisticated robot recently developed by Schindler Group has been executed.

• Science of Emotion and Sensibility
• /
• v.24 no.3
• /
• pp.115-124
• /
• 2021

Non-verbal communication is important in human interaction. It provides a layer of information that complements the message being transmitted. This type of information is not limited to human speakers. In human-robot communication, increasing the animacy of the robotic agent-by using non-verbal cues-can aid the expression of abstract concepts such as emotions. Considering the physical limitations of artificial agents, robots can use light and movement to express equivalent emotional feedback. This study analyzes the effects of LED and motion animation of a spherical robot on the emotion being expressed by the robot. A within-subjects experiment was conducted at the University of Tsukuba where participants were asked to rate 28 video samples of a robot interacting with a person. The robot displayed different motions with and without light animations. The results indicated that adding LED animations changes the emotional impression of the robot for valence, arousal, and dominance dimensions. Furthermore, people associated various situations according to the robot's behavior. These stimuli can be used to modulate the intensity of the emotion being expressed and enhance the interaction experience. This paper facilitates the possibility of designing more affective robots in the future, using simple feedback.

• The Journal of Korean Association of Computer Education
• /
• v.18 no.2
• /
• pp.21-33
• /
• 2015

This paper aims at developing IT convergence robot education contents using open hardware-based GoGo Board and presenting three cases that were applied into educational settings with elementary and middle school students. Several types of data for their activities were collected: photos, work output, survey data, video data and interview with robot teacher and students. Each student experienced building up a GoGo Bumper Car with touch sensors attached at front and back sides and figuring out the principle of digital board control and operating of electronic devices by sensing. The participants, in the following phases, conducted domino chain-reaction with GoGo Bumper Cars and acquiring GoGo Driving Licence by driving test on three different road maps. Students in a gifted education program creatively implemented their own ideas as part of robotic art. The result of case analysis showed that the proposed project provides students not only intimacy for technology, fun, concentration but her own empowerment for developing ideas and creative implementation.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.8
• /
• pp.883-892
• /
• 2018

Recently, as interest in social robots to support physical convenience and emotional sympathy has increased, and social internet has developed, a social robot has evolved as various services simply beyond robot function. Therefore, to develop a social robot service effectively, it is required to study the functional application and methods of interaction between user and social robot service. The purpose of this study is to classify social robot services and to suggest the types of elements that need to be considered in service development. To do this, we conducted in-depth case studies and analysis based on the theoretical definitions and characteristics of social robots. Then, based on the sympathy and functions, we classified social robot services into 1) emotional support type, 2) companion type, 3) guide type, and 4) life support type. In addition, in this study, we derive the considerable factors according to the classified types for the development of effective social robot services. This study will contribute to the understanding and development of various services using a social robot.