• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.92-99
• /
• 2023

Autonomous vehicles should recognize and respond to the specified speed to drive in compliance with regulations. To recognize the specified speed, the most representative method is to read the numbers of the signs by recognizing the speed signs in the front camera image. This study proposes a method that utilizes YOLO-Labeling-Labeling-EfficientNet. The sign box is first recognized with YOLO, and the numeric digit is extracted according to the pixel value from the recognized box through two labeling stages. After that, the number of each digit is recognized using EfficientNet (CNN) learned with the virtual environment dataset produced directly. In addition, we estimated the depth of information from the height value of the recognized sign through regression analysis. We verified the proposed algorithm using the virtual racing environment and GTSRB, and proved its real-time performance and efficient recognition performance.

• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.100-106
• /
• 2023

We developed a crawler-based driving platform that can perform harvesting, transportation, pest control, and rotary operation by equipping it with various implements, and analyzed its performance. This single platform was developed to perform as pepper harvester, peanut harvester, and transporter with a 46-kW engine. A simulation model was developed to study the specifications of the platform, and the accuracy was also analyzed. The absolute percentage error ranged from 0.2 to 5.9%, which made it possible to predict the platform performance using simulation model. In T-test, both torque and speed on field and asphalt showed a significant difference (1%). Driving torque required differed depending on the nature of the field, and the speeds also changed based on soil load. The developed platform has the advantage of being equipped with a variety of working tools, expected to be used to harvest root crops in the future.

• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.107-114
• /
• 2023

Garlic is one of the major seasoned vegetables in Korea along with peppers and onions. Although it is a major cash crop, the aging population of farmers and rising labor and production costs are cited as the reasons for the decreasing production. Accordingly, it is necessary to introduce and spread garlic cultivation mechanization technology to reduce the input labor demand. In this study, conducted a field demonstration in Jeollanam-do using garlic harvesting machinery. In addition, it was intended to improve the mechanization rate of garlic harvesting by deriving factors that can apply to garlic harvesting machines in Jeollanam-do and investigating regional adaptability. As a result of the analysis, it was found that the harvesting performance of the garlic harvester and garlic collector in Jeollanam-do met the agricultural machine test standards. In addition, as a result of calculating the input effect of the mechanized work system compared to the conventional garlic harvesting work system, it was found that there was a labor-saving effect of 96.02%.

• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.133-139
• /
• 2023

A weighing system calculates the bucket's excavation amount of an excavator. Usually, the excavation amount is computed by the excavator's motion equations with sensing data. But these motion equations have computing errors that are induced by assumptions to the linear systems and identification of the equation's parameters. To reduce computing errors, some commercial weighing system incorporates particular motion into the excavation process. This study introduces a linear regression model on an artificial neural network that has fewer predicted errors and doesn't need a particular pose during an excavation. Time serial data were gathered from a 30tons excavator's loading test. Then these data were preprocessed to be adjusted by MPL (Multi Layer Perceptron) or CNN (Convolutional Neural Network) based linear regression models. Each model was trained by changing hyperparameter such as layer or node numbers, drop-out rate, and kernel size. Finally ID-CNN-based linear regression model was selected.

• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.140-149
• /
• 2023

Wire harness is a component for efficient control when electronic parts are required such as construction machinery and electric vehicles. With emerging issues such as autonomous driving and automation in construction, a wire harness composed of multiple cables has become an essential part because more electronic parts are required. However, when a wire harness failure occurs, systems can be stopped, accidents can occur, and economic damage can be significant. Therefore, in this paper, we developed a digital fault screening system that could easily and quickly diagnose faults in the wire harness. The principle of the developed system was to sequentially send pulse signals to the wire harness and use returned signals to perform fault detection. As a result of diagnosing faults using the developed failure detection system, a detection accuracy of 99.9 % was confirmed through the experiments.

• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.9-16
• /
• 2023

This paper analyzed a drainage tower used to drain water in flooded areas. Multi-body dynamics simulation was used to analyze the dynamic behavior of the drainage tower. Structural analysis, flexible-body dynamic analysis, and rigid body dynamic analysis were done to study the maximum Von-Mises stress of the drainage tower. The results showed that the maximum Von-Mises stress occurs at the turn table, and it decreases when the angle of the boom is increased. Also, the rate of the change of angle affects the maximum stress so that the maximum stress changes more when the angular velocity of the boom increases. Based on the rigid body dynamic analysis and the theoretical analysis results, the centrifugal force from the angular velocity makes the difference in the maximum stress at the turn table because of the difference in their direction. Consequently, it was concluded that the centrifugal force should be considered when designing construction machinerythat can rotate.

• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.17-26
• /
• 2023

A forklift is an industrial vehicle that lifts or transports heavy objects using a hydraulically operated fork, and is equipped with an automatic transmission for the convenience of repetitive transportation, loading, and unloading work. The Transmission Control Unit (TCU) is a key component in charge of the shift control function of an automatic transmission. It consists of an electric circuit with an input/output signal interface function and firmware running on a microcontroller. To develop TCU firmware, the development process of shifting algorithm design, firmware programming, verification test, and performance improvement must be repeated. A simulator is a device that simulates a mechanical system having dynamic characteristics in real time and simulates various sensor signals installed in the system. The embedded transmission simulator is a simulator that is embedded in the TCU firmware. information related to the mechanical system that is necessary for TCU normal operation. In this study, an embedded transmission simulator applied to the originally developed forklift TCU firmware was designed and used to verify various forklift shift control algorithms.

• Journal of Drive and Control
• /
• v.21 no.1
• /
• pp.1-7
• /
• 2024

The locking performance of a multi-plate clutch with a mechanical lock-up system is governed by the engagement algorithm. In this paper, a control algorithm to improve the locking performance of the clutch was studied. A 1D dynamic model was constructed and simulated according to the developed algorithm. The developed algorithm was composed of a method in which the locking device is engaged while generating artificial slip on the friction plate by controlling the piston pressure of the clutch. Furthermore, a case study of the parameters within the developed algorithm was conducted to explore combinations that maximize locking performance and analyze trends according to these parameters.

• Journal of Drive and Control
• /
• v.21 no.1
• /
• pp.31-37
• /
• 2024

In this study, to prevent accidents in underground facilities during excavation, we developed a Lv.3 automated control system that can be configured as an electronic control system without changing the existing hydraulic system in a general excavator and utilized digital map information of underground facilities. We aimed to develop a strategy to prevent accidents caused by operator error. To implement this, a real-time excavator bucket end position recognition and control system was developed through angle measurement of the boom, arm, and bucket using an electronic joystick, RTK-GPS, and angle sensors. In addition, excavators are large, machine-based equipment, and it is difficult to control overshoot due to inertia with feedback control using position recognition information of the bucket tip. Therefore, feed-forward control is used to calculate the moving speed of the bucket tip in real-time to determine the target position. We developed a technology that can converge and verified the performance of the developed system through actual vehicle installation and field tests.

• Journal of Drive and Control
• /
• v.21 no.1
• /
• pp.46-52
• /
• 2024

The advancement of autonomous driving technology has heightened the importance of Autonomous Mobile Robotics (AMR) within smart factories. Notably, in tasks involving the transportation of heavy objects, the consideration of weight in route optimization and path planning has become crucial. There is ongoing research on local path planning, such as Dijkstra, A^*, and RRT^*, focusing on minimizing travel time and distance within smart factory warehouses. Additionally, there are ongoing simultaneous studies on route optimization, including TSP algorithms for various path explorations and on minimizing energy consumption in mobile robotics operations. However, previous studies have often overlooked the weight of the objects being transported, emphasizing only minimal travel time or distance. Therefore, this research proposes route planning that accounts for the maximum payload capacity of mobile robotics and offers load-optimized path planning for multi-destination transportation. Considering the load, a genetic algorithm with the objectives of minimizing both travel time and distance, as well as energy consumption is employed. This approach is expected to enhance the efficiency of mobility within smart factories.