Mohamad Hazwan Mohd Ghazali

The quest to identify the safest and most efficient routes for autonomous vehicles through path planning techniques is a captivating challenge. Among these, grid-based mapping methods stand out for their ability to simplify navigation, particularly in complex settings. This paper presents a highly innovative approach for optimization in grid environments, termed modified ant colony optimization (MACO), which enhances traditional ant colony optimization (ACO) for more effective path planning. In this approach, ACO is initially used to explore potential routes within a grid. The core innovation of MACO is its unique probabilistic prediction mechanism that integrates heuristic cues with stochastic evaluation, thereby refining node selection and robustly guiding the search process. This not only improves the chances of finding feasible routes but also boosts success rates across various test environments. Furthermore, a path reconnection process is introduced to ensure both optimal path length and smoothness. The method is also extended to tackle the path planning problem for multiple mobile robots. To achieve this, a new term is added to the objective function to evaluate collisions between robots, and a collision removal operator is applied to resolve potential path conflicts. MACO was tested on nine grid models of varying complexity, showing superior performance compared to other algorithms. It achieved the shortest path and fastest computation times across multiple environments, with a 6% improvement in path length and execution times as low as 0.3 s, even in complex scenarios. Additionally, MACO achieved a 100% success rate in several challenging environments.

Accurate recognition of complex human activities from wearable sensors plays a critical role in sports analytics and human performance monitoring. However, the high dimensionality and redundancy of raw inertial data can hinder model performance and interpretability. This study proposes a hybrid feature selection framework that combines Minimum Redundancy Maximum Relevance (MRMR) and Regularized Neighborhood Component Analysis (RNCA) to improve classification accuracy while reducing computational complexity. Multi-sensor inertial data were collected from field hockey players performing six activity types. Time- and frequency-domain features were extracted from four body-mounted inertial measurement units (IMUs), resulting in 432 initial features. MRMR, combined with Pearson correlation filtering (|ρ| > 0.7), eliminated redundant features, and RNCA further refined the subset by learning supervised feature weights. The final model achieved a test accuracy of 92.82% and F1-score of 86.91% using only 83 features, surpassing the MRMR-only configuration and slightly outperforming the full feature set. This performance was supported by reduced training time, improved confusion matrix profiles, and enhanced class separability in PCA and t-SNE visualizations. These results demonstrate the effectiveness of the proposed two-stage feature selection method in optimizing classification performance while enhancing model efficiency and interpretability for real-time human activity recognition systems.

This paper proposes a novel intelligent switching control strategy for a five-bar lower limb exoskeleton. First, during the support phase, terminal sliding mode control (TSMC) is employed to ensure robust stability and high-torque amplification capabilities. Then, during the swing phase, a hybrid controller combining proportional-integral-derivative (PID) control and the adaptive neuro-fuzzy inference system (ANFIS) is implemented to generate natural and compliant leg movements. Finally, to achieve smooth transitions between phases, an intelligent switching algorithm based on multi-sensor information fusion is proposed. Simulation results demonstrate that the proposed strategy keeps trajectory tracking errors below 0.05 across all gait phases and achieves stable torque amplification ratios ranging from 1:6 to 1:10. This performance significantly reduces the user’s physical exertion. These findings validate the effectiveness of this control framework in improving the stability and comfort of human–machine interaction.

Path planning in robotics and automation often demands solutions that effectively balance efficiency and path smoothness, particularly in diverse and large-scale environments. To address this, this paper presents the Rapidly Exploring Random Tree-Modified Whale Optimization Algorithm II (RRT-MWOAII), an enhanced version of the previous Rapidly Exploring Random Tree-Modified Whale Optimization Algorithm (RRT-MWOA) approach. The RRT-MWOAII is a hybrid technique that combines the strengths of the RRT path planning method and the MWOAII. It utilizes the RRT path as the initial solution and then applies the MWOAII to improve the smoothness of the RRT path. The key advancement in the RRT-MWOAII algorithm is the incorporation of a targeted global search, where the weakest solution is targeted instead of a random one during the global search phase. This targeted approach helps to efficiently identify and refine the weakest link in the solution, leading to improved overall path optimization. Additionally, the algorithm further refines its local solution by employing a neighborhood search, which allows for fine-tuning of the path. Simulations were conducted and evaluated on seven benchmarks, including RRT, RRT*, Bidirectional RRT (BiRRT), and state-of-the-art optimization methods such as the original Whale Optimization Algorithm (WOA), Improved Whale Optimization Algorithm (IWOA) and Sparrow Search Algorithm (SpSA). Experimental results show that MWOAII consistently produced shorter and smoother paths, outperforming its predecessor with a 31.6% improvement in the initial RRT path smoothness and a 21.05% better average smooth cost across all nine maps.

Unmanned aerial vehicles (UAVs) are prone to crashes due to user inexperience, technical issues, or adverse weather conditions. Most commercial UAVs have a real-time monitoring feature, where important UAV parameters can be monitored. However, low-cost, non-ready-to-fly UAVs often lack these capabilities, making assessing their condition challenging or taking preventive measures when faults occur. This paper presents a novel return-to-home (RTH) prediction algorithm for UAVs, which is triggered when a fault is detected in the UAV. The key contribution of this study is the innovative application of LoRa data in the design of the prediction algorithm, which sets it apart from existing reviews and research that have not explored this approach. The data from the LoRa wireless communication network will be utilized in this algorithm, which consists of three critical parameters: the speed of the UAV (V), the flight range (R), and the battery level (T). Specifically, the algorithm utilizes LoRa’s received signal strength indicator (RSSI) data to estimate the flight range and is designed for use within a 1 km flight radius. It provides actionable recommendations to UAV pilots to return the UAV to its home position or to land immediately, which can be accessed through a mobile application. This approach enhances safety by reducing the risk of UAV crashes and ensuring timely interventions. ABSTRAK: Kenderaan udara tanpa pemandu (UAV) terdedah kepada kemalangan kerana faktor seperti pengguna kurang berpengalaman, isu teknikal atau keadaan cuaca buruk. Kebanyakan UAV komersial yang terdapat di pasaran didatangi dengan ciri pemantauan masa sebenar, di mana parameter penting UAV boleh dipantau. Walau bagaimanapun, UAV berkos rendah dan tidak sedia terbang sering kekurangan keupayaan ini, menjadikannya mencabar dalam menilai keadaan atau mengambil langkah pencegahan apabila berlaku kerosakan. Kajian ini mengkaji algoritma baharu ramalan pulang ke pangkalan (RTH) untuk UAV, tercetus apabila kerosakan dikesan dalam UAV. Sumbangan utama kajian ini adalah aplikasi inovatif data LoRa dalam reka bentuk algoritma ramalan, membezakannya daripada kajian terdahulu yang belum menggunakan LoRa dalam konteks ini. Data dari rangkaian komunikasi tanpa wayar LoRa digunakan dalam algoritma ini, iaitu terdiri daripada tiga parameter penting: kelajuan UAV (V), penerbangan (R), dan paras bateri (T). Secara khusus, algoritma menggunakan data penunjuk kekuatan isyarat (RSSI) yang diterima LoRa bagi menganggarkan julat penerbangan dan direka bentuk bagi kegunaan radius penerbangan 1 km. Kajian ini memberi cadangan tindakan yang boleh diambil oleh juruterbang UAV, sama ada bagi mengembalikan UAV ke kedudukan asal atau mendarat serta-merta, yang boleh diakses melalui aplikasi mudah alih. Pendekatan ini meningkatkan keselamatan dengan mengurangkan risiko kemalangan UAV dan memastikan campur tangan pengguna tepat pada masanya.

Advancements in marine vessel stabilization technologies have greatly improved the safety and comfort for crews, cargo, and passengers, while also lowering the risk of capsizing. Marine vessels operating in open waters experience disturbances from unpredictable waves and strong winds, leading to instability and potential accidents. To mitigate these risks, stabilizers are integrated into ships to maintain stability, safety, and seaworthiness. As modern vessels evolve, stabilizers play an increasingly vital role, with gyro stabilizers emerging as a highly effective solution for mitigating roll motion without external appendages. This paper provides a comprehensive review of gyro stabilizer designs and techniques, covering their historical development, mathematical modeling, advantages, limitations, and case studies. A review of current research highlights a strong emphasis on simulating gyro stabilizer control systems, demonstrating their potential in enhancing vessel stability. However, key challenges persist, including accurate wave moment prediction during the design phase and gyroscopic drift, which can impact long-term performance and efficiency. Despite these challenges, recent advancements in control algorithms and optimization techniques have improved gyro stabilizer adaptability under varying sea conditions. By exploring the latest developments, this review serves as a valuable reference for researchers, engineers, and maritime professionals seeking to advance marine stabilization technologies. The findings underscore the potential of next-generation gyro stabilizers, including fiber optic gyros, in achieving higher precision, stability, and operational efficiency.

Unmanned aerial vehicles (UAVs) or UAVs are increasingly applied by military and civilians in a wide range of applications, such as environmental monitoring, search and rescue, target detection, combat, precision agriculture, and three-dimensional (3-D) mapping. The failure of UAVs can cause human casualties and property damage. Hence, it is crucial to consistently monitor the UAV in order to detect any potential malfunction in vital components like motors and electronic speed controllers (ESCs). This study proposed a new UAV condition-monitoring approach that combines sensor fusion and fuzzy-based decision-making algorithms. The vibration, the motor's rotational speed, and current parameters are utilized to determine the UAV's condition, specifically the UAV's motors or ESCs. The UAV's condition is categorized into safe, partial safe, and danger. Experimental results show a clear distinction between a healthy and faulty UAV's component in terms of vibration, with a discrepancy of more than 100% in the majority of the cases. The proposed framework can successfully alert the user if there is a malfunction in the UAV's motor or ESC before or when the UAV is flying.

Early monitoring of the drone’s critical components, such as the motor, propeller, and electronic speed controller (ESC), can prevent unwanted accidents from happening to the drone. A failure of one of the drone’s components may lead the drone to crash. The fault detection approach can be made by analyzing certain parameters such as vibration, acoustic, current, and speed. In this paper, a comparative analysis between data-driven and visual-based approaches in detecting an imbalanced propeller in the drone is presented. The analysis will focus on vibration and acoustic parameters, with the use of CZN-15 and ADXL335 MEMS sensors to capture vibration and acoustic data, respectively. The Silky EvCam vibration camera will be used for the visual-based approach. Experimental results demonstrated that both data-driven and visual-based approaches could detect the imbalanced UAV propeller. There is a shift of vibration frequency when imbalances occur in the UAV propeller. However, in the ground and thrust modes, the acoustic-based method has an accuracy of 54% and 93%, respectively, which is lower compared to the vibration-based (100% accuracy in both modes) and visual-based techniques.