Current Issue : October-December Volume : 2026 Issue Number : 4 Articles : 5 Articles
Controller Area Network (CAN) remains a key in-vehicle communication protocol for distributed automotive control systems, where predictable communication timing is essential for coordinated operation of electronic control units (ECUs). This paper presents a cross-validated framework for timing analysis of automotive CAN networks by combining Deterministic and Stochastic Petri net (DSPN) modeling with worst-case response-time (WCRT) analysis. A DSPN model is developed to represent CAN message generation, priority-based arbitration, bus access, and non-preemptive frame transmission. The model is implemented in TimeNet to evaluate bus utilization, queue occupancy, and access-delay behavior under representative automotive traffic. In parallel, analytical WCRT equations are used to derive conservative latency bounds for each message class. The proposed framework links stochastic performance observations from DSPN simulation with deterministic schedulability guarantees from WCRT analysis, enabling consistency checks between average-case and worst-case timing results. A case study based on a 500 kbit/s automotive CAN configuration with six priority classes is presented. The results show that the network operates at approximately 35.9% bus utilization and that all message classes satisfy their timing requirements with a substantial margin, with the maximum worst-case response time remaining below 2 ms. The study further discusses the modeling assumptions, abstraction limits, and sensitivity of timing behavior to frame length and traffic configuration. The proposed framework provides a practical methodology for timing-oriented design and early-stage validation of automotive CAN communication systems....
The control of complex nonlinear robotic systems is often considered computationally intensive, typically relying on 32-bit or more powerful hardware. This work challenges that paradigm by presenting a novel implementation of a fully data-driven intelligent controller on a resource-constrained microcontroller for robotics. A ball-and-plate system coupled to a six degree-of-freedom Stewart–Gough parallel manipulator, a widely used system for evaluating nonlinear control strategies, was controlled using an artificial neural network (ANN) deployed on an Arduino UNO R3. The datasets were created through dedicated experiments, comprising over 5 h of operation, 1110 repetitions, and 243,000 data points which included key parameters such as target and measured position, elapsed time, servomotor angles, and program flags generated by a baseline PID controller. The ANN architecture featured six inputs, one hidden layer with 12 neurons, and six outputs, and was ported to the 8-bit Arduino UNO using the AIfES (Artificial Intelligence for Embedded Systems) framework. Comparative experiments against proportional-integral-derivative (PID) control demonstrated that, while maintaining comparable performance in nominal conditions, the ANN controller exhibited superior stability and robustness under nonlinear and extreme scenarios, especially under increased speeds that break the linearized PID assumptions, making it unstable. The ANN also demonstrated the ability to incorporate actuation saturation into the control model. These results demonstrate the superior robustness of intelligent controllers for nonlinear control, while highlighting the feasibility of deploying data-driven intelligent controllers on ultra-low-power embedded devices, broadening the application space of Tiny Machine Learning (TinyML) in real-time robotics....
In the past few decades, researchers have developed various pressure management techniques to control pressure, recover energy, and reduce leakage in water distribution networks. Most of these techniques have been developed and validated under numerical conditions. Only rarely has it been possible to receive feedback from the field after application, creating a gap between the numerical approach and field application. To address this gap, the Experimental Network for the Evaluation of the Digital Twin (E-NET) has been developing at Politecnico di Milano as a scaled laboratory model of a widespread numerical benchmark network to create a bridge between numerical simulations and laboratory practice. In this study, a well-known technique for Pressure Reducing Valves (PRVs) placement was applied to both the original benchmark and the scaled laboratory (E-NET) networks. It was confirmed that leakage reduction, as well as the optimal location and setting of PRVs, remain consistent across both systems. This demonstrates the feasibility of developing methodologies based on the scaled benchmark, thereby opening the way for experimental verification of the proposed methods....
Although model predictive control (MPC) has been successfully applied in permanent magnet synchronous motor (PMSM) speed control systems, its performance can degrade under high-dynamic operating conditions and uncertain load disturbances. To address these issues, a continuous-time model predictive control (CTMPC) framework is proposed to improve speed tracking accuracy and robustness. From a symmetry perspective, the proposed method leverages the orthogonal symmetry of Laguerre basis functions and the structural invariance of the continuous-time PMSM speed dynamics, enabling a compact and balanced representation of the control trajectory while preserving prediction accuracy. Specifically, a finite set of orthogonal Laguerre functions, combined with an adaptive smoothing factor and soft constraint mechanism, is employed to reduce computational complexity without compromising performance. In addition, a nonlinear disturbance observer is integrated to achieve real-time estimation and feedforward compensation of load torque variations, thereby enhancing disturbance rejection capability. Comprehensive simulation results demonstrate that the proposed approach significantly improves tracking precision, reduces overshoot, and shortens recovery time following load disturbances compared to conventional MPC methods....
The of personal computer (PC) tasks represents a systems-level challenge that integrates natural language processing, visual perception and mouse–keyboard action control. While existing approaches mainly focus on the application programming interface (API)-based or terminal-based automation, which are incompatible with the majority of applications for the lack of accessible interface. In this article, we propose PCLLM, a novel end-to-end system that automates PC operations by integrating large language models (LLMs) with computer vision techniques to directly control the mouse and keyboard. First, a software knowledgebased prompt engineering method is developed to comprehend software architecture and operational sequences. Second, template matching techniques are integrated for precise element localization, allowing the system to accurately identify and interact. Third, a dual- LLM pipeline is designed to automatically generate the test data, where a questioner LLM generates diverse task commands and the PCLLM executes these tasks, the corresponding process data are recorded automatically for performance evaluation. Finally, PCLLM is further validated through three typically PC applications (Notepad, Wordpad and Calculator), demonstrating its flexible and robust performance towards intelligent PC automation. To evaluate the proposed system, we adopt task completion rate as the primary metric. Experimental results show that PCLLM achieves the highest completion rates of 98.59%, 95.77%, and 52.11% on Notepad for basic, intermediate, and advanced tasks respectively when powered by GPT-4o, outperforming the CogAgent baseline. These results demonstrate the effectiveness of our approach for PC task automation....
Loading....