• Korean Journal of Agricultural Science
• /
• v.47 no.2
• /
• pp.205-217
• /
• 2020

In this study, a deep learning-based tillage boundary detection method for autonomous tillage by a tractor was developed, which consisted of image cropping, object classification, area segmentation, and boundary detection methods. Full HD (1920 × 1080) images were obtained using a RGB camera installed on the hood of a tractor and were cropped to 112 × 112 size images to generate a dataset for training the classification model. The classification model was constructed based on convolutional neural networks, and the path boundary was detected using a probability map, which was generated by the integration of softmax outputs. The results show that the F1-score of the classification was approximately 0.91, and it had a similar performance as the deep learning-based classification task in the agriculture field. The path boundary was determined with edge detection and the Hough transform, and it was compared to the actual path boundary. The average lateral error was approximately 11.4 cm, and the average angle error was approximately 8.9°. The proposed technique can perform as well as other approaches; however, it only needs low cost memory to execute the process unlike other deep learning-based approaches. It is possible that an autonomous farm robot can be easily developed with this proposed technique using a simple hardware configuration.

• Journal of The Korean Association of Information Education
• /
• v.23 no.5
• /
• pp.441-449
• /
• 2019

With the introduction of the 2015 revised curriculum, software education became mandatory in elementary school. The practical subject in the content for the software education consists of information ethics, algorithms, programming, and computer science. Especially, elementary school encourages the unplugged activities. Unplugged activities are preferred among the teachers in the education field as a teaching-learning method. However, those teachers pointed out that the lack of suitable unplugged activity materials was the biggest challenge. In addition, it was pointed out that the existing materials were not suitable for achievement standards, and there are many simple playing-oriented educations that are not linked to computing activities. In this study, we developed an unplugged activity-oriented robot that can be used for the elementary students to learn the basic control structure suggested in the achievement criteria SW education and to develop creative problem solving ability through various activities using sensors. The effectiveness was verified through the software class using the developed robot in this study.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.141-148
• /
• 2022

Recently, reinforcement learning combined with deep neural network technology has achieved remarkable success in various fields such as board games such as Go and chess, computer games such as Atari and StartCraft, and robot object manipulation tasks. However, such deep reinforcement learning describes states, actions, and policies in vector representation. Therefore, the existing deep reinforcement learning has some limitations in generality and interpretability of the learned policy, and it is difficult to effectively incorporate domain knowledge into policy learning. On the other hand, dNL-RRL, a new relational reinforcement learning framework proposed to solve these problems, uses a kind of vector representation for sensor input data and lower-level motion control as in the existing deep reinforcement learning. However, for states, actions, and learned policies, It uses a relational representation with logic predicates and rules. In this paper, we present dNL-RRL-based policy learning for transportation mobile robots in a manufacturing environment. In particular, this study proposes a effective method to utilize the prior domain knowledge of human experts to improve the efficiency of relational reinforcement learning. Through various experiments, we demonstrate the performance improvement of the relational reinforcement learning by using domain knowledge as proposed in this paper.

• Journal of IKEEE
• /
• v.19 no.2
• /
• pp.276-281
• /
• 2015

Engineers increasingly rely on computers and their computer programming skills for their works. As a result, most engineering curricula have introduced a computer programming courses. However, students consider the subject to be unrelated to their core interests and often feel uncomfortable when learning to program for the first time. To overcome these difficulties, several studies have proposed the use of physical computing paradigm. This paradigm takes the computational concepts out of the PC screen and into the real world so that the student can interact with them. This paper proposes Arduino platform as a tool for attracting interest of the programming and reports the results of questionnaire survey analysis.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.3
• /
• pp.264-269
• /
• 2006

This paper presents the strategy for searching a hidden object in an unknown area for using by multiple distributed autonomous robotic systems (DARS). To search the target in Markovian space, DARS should recognize th ε ir surrounding at where they are located and generate some rules to act upon by themselves. First of all, DARS obtain 6-distances from itself to environment by infrared sensor which are hexagonally allocated around itself. Second, it calculates 6-areas with those distances then take an action, i.e., turn and move toward where the widest space will be guaranteed. After the action is taken, the value of Q will be updated by relative formula at the state. We set up an experimental environment with five small mobile robots, obstacles, and a target object, and tried to research for a target object while navigating in a un known hallway where some obstacles were placed. In the end of this paper, we present the results of three algorithms - a random search, an area-based action making process to determine the next action of the robot and hexagon-based Q-learning to enhance the area-based action making process.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.6-8
• /
• 2004

The objective of this paper is to develop a powered wheelchair controller based on EMG for users with high-level spinal cord injury using a proportional control scheme. An advantage of EMG is relative convenience of acquisition by a surface electrode to users. Direction information can be easily extracted from two EMG channels and force information can be acquired by proportional relationship between the amplitude of EMG and user's power, respectively. Pattern classification algorithm is a threshold method with a supervised learning process. Furthermore, the emergency situation can be avoided using an interrupt function. We evaluated the performance of powered wheelchair controller by navigating a pre-defined path with three non-handicapped people. The results show the feasibility of EMG as an input interface for powered wheelchair and other devices for the seriously disabled.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.640-645
• /
• 2004

This paper proposes a direct adaptive control method using self-recurrent wavelet neural network (SRWNN) for stable path tracking of mobile robots. The architecture of the SRWNN is a modified model of the wavelet neural network (WNN). Unlike the WNN, since a mother wavelet layer of the SRWNN is composed of self-feedback neurons, the SRWNN has the ability to store the past information of the wavelet. For this ability of the SRWNN, the SRWNN is used as a controller with simpler structure than the WNN in our on-line control process. The gradient-descent method with adaptive learning rates (ALR) is applied to train the parameters of the SRWNN. The ALR are derived from discrete Lyapunov stability theorem, which are used to guarantee the stable path tracking of mobile robots. Finally, through computer simulations, we demonstrate the effectiveness and stability of the proposed controller.

• The Journal of Korea Robotics Society
• /
• v.12 no.3
• /
• pp.313-321
• /
• 2017

One of the most frequently performed tasks in human-robot interaction (HRI), intelligent vehicles, and security systems is face related applications such as face recognition, facial expression recognition, driver state monitoring, and gaze estimation. In these applications, accurate head pose estimation is an important issue. However, conventional methods have been lacking in accuracy, robustness or processing speed in practical use. In this paper, we propose a novel method for estimating head pose with a monocular camera. The proposed algorithm is based on a deep neural network for multi-task learning using a small grayscale image. This network jointly detects multi-view faces and estimates head pose in hard environmental conditions such as illumination change and large pose change. The proposed framework quantitatively and qualitatively outperforms the state-of-the-art method with an average head pose mean error of less than $4.5^{\circ}$ in real-time.

• The Journal of Korea Robotics Society
• /
• v.12 no.2
• /
• pp.206-216
• /
• 2017

This paper proposes a method to simultaneously estimate two degrees of freedom in wrist forces (extension - flexion, adduction - abduction) and one degree of freedom in grasping forces using Electromyography (EMG) signals of the forearms. To correlate the EMG signals with the forces, we applied a multi - layer perceptron(MLP), which is a machine learning method, and used the characteristics of the muscles constituting the forearm to generate learning data. Through the experiments, the similarity between the MLP target value and the estimated value was investigated by applying the coefficient of determination ($R^2$) and root mean square error (RMSE) to evaluate the performance of the proposed method. As a result, the $R^2$ values with respect to the wrist flexion-extension, adduction - abduction and grasping forces were 0.79, 0.73 and 0.78 and RMSE were 0.12, 0.17, 0.13 respectively.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.1
• /
• pp.1-11
• /
• 1994

This paper presents an intelligent robotic control method for chamferless parts mating by integrating fuzzy control and neural network. The successful assembly task requires an extremely high position accuracy and a good knowledge of mating parts. However, conventional assembly method alone makes it difficult to achieve satisfactory assembly performance because of the complexity and the uncertainties of the process and its environments such as not only the limitation of the devices performing the assembly but also imperfect knowledge of the parts being assembled. To cope with these problems, an intelligent robotic assembly method is proposed, which is composed of fuzzy controller and learning mechanism based upon neural net. In this method, fuzzy controller copes with the complexity and the uncertainties of the assembly process, while neural network enhances the assembly scheme so as to learn fuzzy rules from experience and adapt to changes in environment of uncertainty and imprecision. The performance of the proposed assembly scheme is evaluted through a series of experiments using SCARA robot. The results show that the proposed control method can be effectively applied to chamferless precision parts mating.