Ali Rospawan

Ali Rospawan

Auto-generated profile, based on publicly available information

Ali Rospawan

Department of Electrical Engineering, National Chung Hsing University, Taichung, Taiwan

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 January 2024 in IEEE Access
Verified
Journal Article
Open access
ARAli Rospawan
Ali Rospawan
CTChing-Chih Tsai
Ching-Chih Tsai
CHChi-Chih Hung
Chi-Chih Hung
This paper proposes a novel adaptive predictive Proportional-Integral-Derivative (PID) controller utilizing an output recurrent fuzzy broad learning systems (ORFBLS) for Multiple-Input Multiple-Output (MIMO) digital control systems, aiming to effectively adapt to changing setpoints and dynamic environments. The proposed controller, MIMO ORFBLS-APPID controller in short, is proposed to extend the application of ORFBLS as an adaptive adjustment mechanism for PID gains parameters, where the controller gain matrices are automatically tuned over time by employing the Jacobian transformations of the MIMO ORFBLS identifier. Three theorems are established to ensure proper usage and successful applications of the proposed controller. The setpoints tracking control performance and disturbance rejection abilities are firmly illustrated by performing three simulations to the multivariable nonlinear dynamic systems. Moreover, one experimental study to the laboratory-built extrusion barrel in a plastic injection molding machine is done to validate the effectiveness and practicality of the proposed control method. Through comparative simulations and experimental results, the proposed controller has been shown to outperform two existing control methods in terms of control performance indexes.

24 June 2024 in International Journal of Fuzzy Systems
Verified
Journal Article
Closed access
PDF file
ARAli Rospawan
Ali Rospawan
CTChing-Chih Tsai
Ching-Chih Tsai
CHChi-Chih Hung
Chi-Chih Hung

20 February 2025 in IEEE Access
Verified
Journal Article
Open access
ARAli Rospawan
Ali Rospawan
CTChing-Chih Tsai
Ching-Chih Tsai
CHChi-Chih Hung
Chi-Chih Hung
This paper presents a novel two-layer intelligent learning control framework using an output recurrent fuzzy neural LSTM broad learning system (ORFNLSTM-BLS) and a RMSprop optimization algorithm, designed to enhance learning capability and address the limitations of existing fuzzy BLS approaches. The first layer focuses on real-time system identification, leveraging ORFNLSTM-BLS to learn complex nonlinear system dynamics and effectively capture long-term dependencies in sequential data. The second layer implements an intelligent adaptive controller based on the ORFNLSTM-BLS, which dynamically adjusts control actions for time-varying setpoint tracking and disturbance rejection. The RMSprop optimization algorithm is applied across both layers to adaptively regulate learning rates, ensuring faster convergence and mitigating gradient issues. The convergence of the proposed method is provided through five theorems, ensuring gradient boundedness and learning rate conditions. Additionally, a comprehensive study on time complexity and memory consumption is presented to guide controller board selection. Simulations and experimental results demonstrate the superiority of the proposed framework in achieving accurate control performance, robustness under uncertainty, and real-world applicability, where conventional PID control fails to meet the required system response. This method offers valuable insights for developing advanced intelligent control solutions in dynamic environments involving sequential data and uncertainty.

17 November 2022 in Jurnal ELTIKOM
Verified
Journal Article
Open access
PDF file
ARAli Rospawan
Ali Rospawan
JSJoni Welman Simatupang
Joni Welman Simatupang
IPIrwan Purnama
Irwan Purnama
In industrial application on machine manufacturer multinational company, especially on the conveyor manufacturing company, before delivering the conveyor to overseas customer, there are some regulations need to be complied to ensure the product quality is always good. The system to ensure the product quality before delivery is named by Factory Acceptance Test (FAT). The purpose of doing this is to gain the trust and satisfaction of the buyer company by ensuring the quality before delivery. To ensure the product quality, this device was designed following the customer check sheet and requirement by giving them reliable data of the load current on each line of the motor and the motor temperature. This device is consisting of WEMOS D1 mini as the main controller and an IoT device, thermocouple type K as a temperature sensor, MAX6675 as the driver of the thermocouple, SCT-013-005 as a split-core current sensor, Arduino Nano as the second controller, and LCD 2004 as a real-time display. The IoT Application for Conveyor Motor Load Current and Temperature Monitoring Device is successfully created with the output is a table of data and the line chart which consists of Load current on each phase (U1, V1, W1), the motor temperature, and the limitation of each data so the user is easier to determine the commissioning result of the Factory Acceptance Test.

17 February 2026 in International Journal of Fuzzy Systems
Verified
Journal Article
Closed access
CTChing-Chih Tsai
Ching-Chih Tsai
GYGuo-Hwa Yang
Guo-Hwa Yang
SCShih-Che Chen
Shih-Che Chen
ARAli Rospawan
Ali Rospawan
HZHong-Yu Zheng
Hong-Yu Zheng

27 March 2026 in IEEE Sensors Journal
Verified
Journal Article
Closed access
CTChing-Chih Tsai
Ching-Chih Tsai
ARAli Rospawan
Ali Rospawan
CLChi-Hsiang Li
Chi-Hsiang Li
HLHung-Sheng Lin
Hung-Sheng Lin
SCShih-Che Chen
Shih-Che Chen
The original ORB-SLAM3 has been designed with static backgrounds in mind. However, when dealing with moving obstacles in a dynamic environment, this method could lead to significant localization errors and inaccuracies in the generated maps. This paper presents an enhanced ORB-SLAM3 method to mitigate these errors caused by dynamic environments and moving obstacles. To achieve this, we integrate the known Photogrammetric Key-Frame Selection (PKS) algorithm to optimize keyframe selection, and a Static Conditional Probability Model (SCPM) for effective identification and eliminating dynamic map points in each frame. Extensive evaluation simulations demonstrate that our approach substantially reduces Absolute Trajectory Error (ATE) metrics compared to the original ORB-SLAM3, with details as follows; mean ATE by 84% on TUM datasets and 17.91% in Gazebo simulations, median ATE by 86% and 28.19%, root mean squared error (RMSE) of ATE by 80% and 29.45%, and standard deviation (STD) of ATE by 71.5% and 61.73%, respectively. This performance improvement comes with a modest computational cost increase of 12.5% on TUM datasets and 31% in Gazebo environments. Furthermore, the experimental validation with an autonomous mobile robot (AMR) in dynamic indoor environments further demonstrates that the proposed enhanced ORB-SLAM3 method achieves superior accuracy, performance, and robustness, with 51.56%, 33.49%, 27.46%, and 9.99% of improvement in terms of mean, median, RMSE and STD of ATE, making it particularly valuable for real-world robotic applications in crowded settings.