Journal of Advances in Artificial Life Robotics Volume 4 Issue3

Volumje 4, Issue 3 December 2023

Research Article
1.Realizing Variable Stiffness through Structures Designed for Quick Motion
Katsuaki Suzuki¹, Yuya Nishida², Kazuo Ishii^2
1Kumamoto Industrial Research Institute, 3-11-38 Higashi machi, Higashi-ku, Kumamoto 862-0901, Kumamoto, Japan
²Life Science and Systems Engineering, Kyushu Institute of Technology, 2-4 Hibikino, Wakamatsu-ku, Kitakyushu 808-0196, Fukuoka, Japan
pp. 122–127
ABSTRACT
To expand a robot’s range of activities, we aim to realize three functions: normal motion, quick motion, and variable stiffness. In this study, we developed a method for realizing variable stiffness using a joint mechanism that can achieve normal and quick motions. Our analysis results confirmed that stiffness could be changed by adjusting the initial displacement of the spring incorporated in the joint mechanism. Furthermore, we manufactured a machine containing the proposed mechanism and evaluated the proposed method by comparing our theoretical and experimental results considering errors. Our findings showed similar trends between the measured and theoretical values with an error caused by friction between machine components.

ARTICLE INFO
Article History
Received 25 November 2022

Accepted 15 November 2023

Keywords
Variable stiffness
Link mechanism

4301

Research Article
2.Automated Food Quality Checking System
Ashraf Ali Jamal Deen¹, Thivagar Chettiar Sarawanan¹, Heshalini Rajagopal², Madam Devika Sethu¹,Norrima Binti Mokhtar³, Neesha Jothi^2
1Department of Electrical and Electronic Engineering, Manipal International University, Malaysia
²Institute of Computer Science and Digital Innovation, UCSI University, 56000 Kuala Lumpur, Malaysia
³Department of Electrical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
pp. 128–132
ABSTRACT
Ensuring food safety, hygiene, and health is crucial in tackling the issue of food wastage. Preserving the high quality of food necessitates vigilant checking and safeguarding against deterioration caused by environmental factors such as temperature, humidity, and light. This research aims to develop a comparable tool for checking food quality, specifically focusing on environmental aspects like temperature, humidity, alcohol content, and light exposure for perishable items like fruits and vegetables. The system is constructed using an Arduino UNO microcontroller, which interfaces with various sensors like DHT-22 to track temperature and humidity, MQ3 to detect alcohol levels, and LDR to measure light exposure. The collected sensor data is transmitted to an IoT platform through an ESP8266 Wi-Fi Module. This IoT system serves to log and monitor the sensor data, enabling remote and real-time oversight of food storage conditions from any location and at any time.

ARTICLE INFO
Article History
Received 06 August 2022
Accepted 25 August 2023

Keywords
Food quality
Quality checking
Internet of things (IoT)
Sensor

4302

Research Article
3.Rescue Robot Challenge to Develop Problem-Solving Skills in Junior High School Students
Kazuo Kawada, Keita Murai, Yuta Susawa, Hiroyuki Y. Suzuki
Graduate School of Humanities and Social Sciences, Hiroshima University, 7398524 Hiroshima, Japan
pp. 133–137
ABSTRACT
The ”Junior High School Rescue Robot Challenge'' is an activity held every year by Hiroshima University in cooperation with construction machinery manufacturers. The challenge has been ongoing for 18 years with the given theme of rescue. However, starting in 2022, the program's policy has essentially changed, with emphasis placed on both problem-finding and problem-solving abilities. A prototype rescue robot was created by modifying a radio-controlled excavator model, and students were responsible for everything from proposing problems to be solved in the disaster area to finding solutions.

ARTICLE INFO
Article History
Received December 7, 2022
Accepted 27 March 2024

Keywords
Rescue robot
Junior high school
Technology education
Robot evaluation
Modified excavator

4303

Research Article
4.Detecting small sample objects with improved YOLOv5
Yuxuan Gao¹, Jiwu Wang¹, Zixin Li^2
1School of Mechanical and Electronic Engineering, Beijing Jiaotong University, Beijing, Haidian District, China
²Aero Engine Corporation of China, Beijing, Changping District, China
pp. 138–141
ABSTRACT
Object detection is widely used in various production and life, such as mask detection and recognition during epidemics, masked face recognition, etc., but object detection algorithms based on deep learning have always been an important research topic in the object detection field. It is an implementation method. Due to the large number of lead seals and fuses, their positions are not fixed, the sample data set is small, the target background is complex and easily occluded, and there are strong reflection interferences, which pose problems with conventional image processing methods. Effectively solving object recognition problems is difficult. In this study, by expanding the dataset, using various data enrichment methods, and training with improved algorithms, the detection accuracy, detection speed, and adaptability were effectively improved.

ARTICLE INFO
Article History
Received December 18, 2022
Accepted 28 March 2024

Keywords
YOLOv5
Deep learning
Object detection
Few-shot learning

4304

Research Article
5.Identifying Industrial Control Systems Anomalies Operation Based on Finite-State Machines
I-Hsien Liu¹, Pei-Wen Chou¹, Nai-Yu Chen¹, Jung-Shian Li^2
1Department of Electrical Engineering / M.S. Degree Program on Cyber-Security Intelligence, National Cheng Kung University, No.1, University Rd., East Dist., Tainan City 701401, Taiwan
²Department of Electrical Engineering / Institute of Computer and Communication Engineering, National Cheng Kung University, No.1, University Rd., East Dist., Tainan City 701401, Taiwan
pp. 142–145
ABSTRACT
During 2021, a facility responsible for treating water in Florida, USA, experienced a cyberattack orchestrated by external malicious entities. These attackers sought to tamper with the levels of certain chemicals in an attempt to compromise the integrity and safety of the water supply. Recognizing the complexity of detecting abnormal operations in Industrial Control Systems (ICS), this research sought to leverage the benefits of finite-state machine technology to address this challenge. To evaluate the scheme, this research performed a series of assessments utilizing the cybersecurity testbed for dam control systems developed by TWISC@NCKU in Taiwan. Research results show that improved detection of irregular operational behaviors, empowering maintenance staff to swiftly pinpoint anomalies.

ARTICLE INFO
Article History
Received 30 November 2023
Accepted 15 May 2024

Keywords
FSM
ICS security
Dam control systems
Anomaly detection

4305

Research Article
6.Robotics Education for Elementary and Junior High School Students Using Beauto Balancer
Youta Takano¹, Kenji Kimura^2
1Belair-systems, Inc, 20-8 Hina-minami-machi, Okazaki-shi, Aichi, 444-0915, Japan
²Department of Control Engineering, National Institute of Technology, Matsue College, 14-4 Nishi-ikuma-cho, Matsue-shi, Shimane,690-8518, Japan
pp. 146–152
ABSTRACT
In recent years, it has become necessary to use programming in controlling robots. However, there are few educational materials for elementary and junior high school students to understand and be interested in the concepts of programming and control engineering. In the previous study, we planned an educational program using "Beauto Balancer2", an educational robot developed by Vstone Corporation, to explain the mechanism of inverted control and to allow students to actually touch the robot, so that they can experience various technologies including control engineering and become interested in mathematical subjects in general, which are the basis of these technologies. The educational program was planned so that students could experience various technologies including control engineering through hands-on experience with robots and develop an interest in mathematical subjects in general. In addition, a craft class was conducted and evaluated using the results of questionnaires. In this study, mock classes were conducted using the existing software for educational robots created in the previous study, and comparisons and verifications were made by focusing on three persons with similar characteristics.

ARTICLE INFO
Article History
Received 30 November 2023
Accepted 17 June 2024

Keywords
Robot education
Control engineering
Visual programming
Additional experiments

4306

Research Article
7.Generation of Labyrinth Images Using Gradients Between Circles in Window
Toru Hiraoka, Kyosuke Matsumoto
Graduate School of Regional Design and Creation, University of Nagasaki, Japan
pp. 153–157
ABSTRACT
Non-photorealistic rendering methods to automatically generate labyrinth images from photographic images have been proposed. Labyrinth images are non-photorealistic images that maze-like intricate lines are laid out at approximately equal intervals on photographic images. Since the shape of the lines forming labyrinth images differs depending on the conventional method, it gives a different impression. We propose a method to generate labyrinth images composed of lines with a shape different from the conventional methods. The proposed method is performed by iterative computation with gradients between circles in a window. To verify the effectiveness of the proposed method, experiments were conducted to visually confirm labyrinth images generated from various photographic images.

ARTICLE INFO
Article History
Received 20 May 2024
Accepted 29 May 2024

Keywords
Non-photorealistic rendering
Labyrinth image
Photographic image
Automatic generation

4307

Research Article
8.GENERATION of Lines-Composed Images that Change Brightness and Width of Lines
Toru Hiraoka, Jia-Lin Zhang
Graduate School of Regional Design and Creation, University of Nagasaki, Japan
pp. 158–162
ABSTRACT
We propose a non-photorealistic rendering method to automatically generate lines-composed images from photographic images. The brightness and width of the lines in lines-composed images change according to the change in the light and dark of photographic images. Our method is executed by repeating a process of updating images using the area near the target pixel that is the most different from the area centered on the target pixel. To verify the effectiveness of our method, we conducted experiments with changing the values of parameters in our method. The experimental results show that our method can generate impressive non-photorealistic images.

ARTICLE INFO
Article History
Received 20 May 2024
Accepted 29 May 2024

Keywords
Non-photorealistic rendering
Line
Brightness
Width
Different area
Automatic generation

4308

Research Article
9.Conceptual Dictionaries for Adjective Concepts and Adjective Verb Concepts in an Integrated Narrative Generation System
Jumpei Ono¹, Takashi Ogata^2
1Faculty of Software and Information Technology, Aomori University, 2-3-1 Kohbata, Aomori, Aomori, 030-0943, Japan
²Faculty of informatics, Yamato University, 2-5-1, Katayama-cho, Suita, Osaka, 564-0082, Japan
pp. 163–171
ABSTRACT
In our narrative generation study, we chiefly use the verb and noun conceptual dictionaries in the narrative generation system. Moreover, other conceptual dictionaries are also necessary for generating more interesting stories. In this study, we present basic methods for developing an adjective conceptual dictionary and adjective verb conceptual dictionary in an integrated narrative generation system (INGS). We refer to antonym relationships between adjective concepts and adjective verb concepts. Furthermore, as an approach to the application based on the above types of dictionaries, we introduce an attempt to use both the description of antonym relationships for the adjective & adjective verb concepts and our “coloring” technique for giving a certain atmosphere to a story to be generated.

ARTICLE INFO
Article History
Received 30 November 2023
Accepted 12 May 2024

Keywords
Adjective conceptual dictionary
Adjective verb conceptual dictionary
Integrated narrative generation
System (INGS)
Antonym
Coloring

4309

Research Article
10.Extending BWDM to Support Various Types and Recursive Definitions in VDM++ Test Case Generation
Shota Takakura¹, Tetsuro Katayama¹, Yoshihiro Kita², Hisaaki Yamaba¹, Kentaro Aburada¹, Naonobu Okazaki^1
1Department of Computer Science and Systems Engineering, Faculty of Engineering, University of Miyazaki, 1-1 Gakuen-kibanadai nishi, Miyazaki, 889-2192 Japan
²Department of Information Security, Faculty of Information Systems, Siebold Campus, University of Nagasaki, 1-1-1 Manabino, Nagayo-cho, Nishi-Sonogi-gun, Nagasaki, 851-2195 Japan
pp. 172–176
ABSTRACT
Generating test cases from VDM++ formal specifications, which help to eliminate ambiguities, is both time-consuming and labor-intensive. To solve this problem, our laboratory has developed BWDM, a tool for automatic test case generation from VDM++ specifications. However, the original BWDM only supports integer types and cannot handle test cases for operations and functions with recursive structures. To enhance BWDM's usefulness, this paper introduces extensions to address these limitations. The results confirm that the extended BWDM can reduce test case generation time compared to manual methods.

ARTICLE INFO
Article History
Received 24 November 2023
Accepted 28 June 2024

Keywords
Software testing
Formal methods
VDM++
Automatic generation
Test cases

4310