Search

The integration of machine learning (ML) into cyber-physical systems (CPS) offers significant benefits, including enhanced efficiency, predictive capabilities, real-time responsiveness, and the enabling of autonomous operations. This convergence has accelerated the development and deployment of a range of real-world applications, such as autonomous vehicles, delivery drones, service robots, and telemedicine procedures. However, the software development life cycle (SDLC) for AI-infused CPS diverges significantly from traditional approaches, featuring data and learning as two critical components. Existing verification and validation techniques are often inadequate for these new paradigms. In this study, we pinpoint the main challenges in ensuring formal safety for learningenabled CPS.We begin by examining testing as the most pragmatic method for verification and validation, summarizing the current state-of-the-art methodologies. Recognizing the limitations in current testing approaches to provide formal safety guarantees, we propose a roadmap to transition from foundational probabilistic testing to a more rigorous approach capable of delivering formal assurance.

Applying high-speed real-time wireless technologies in industrial applications has the great potential to reduce the deployment and maintenance costs compared to their wired counterparts. Wireless technologies enhance the mobility and reduce the communication jitter and delay for mobile industrial equipment, such as mobile collaborative robots. Unfortunately, most existing wireless solutions employed in industrial fields either cannot support the desired high-speed communications or cannot guarantee deterministic, real-time performance. A more recent wireless technology, RT-WiFi, achieves a good balance between high-speed data rates and deterministic communication performance. It is however developed on commercial-of-the-shelf (COTS) hardware, and takes considerable effort and hardware expertise to maintain and upgrade. To address these problems, this paper introduces the software-defined radio (SDR)-based RT-WiFi solution which we call SRT-WiFi. SRT-WiFi provides full-stack configurability for high-speed real-time wireless communications. We present the overall system architecture of SRT-WiFi and discuss its key functions which achieve better timing performance and solve the queue management and rate adaptation issues compared to COTS hardware-based RT-WiFi. To achieve effective network management with rate adaptation in multi-cluster SRT-WiFi, a novel scheduling problem is formulated and an effective algorithm is proposed to solve the problem. A multi-cluster SRT-WiFi testbed is developed to validate the design, and extensive experiments are performed to evaluate the performance at both device and system levels., Comment: 16 pages

Robot programming typically makes use of a set of mechanical skills that is acquired by machine learning. Because there is in general no guarantee that machine learning produces robot programs that are free of surprising behavior, the safe execution of a robot program must utilize monitoring modules that take sensor data as inputs in real time to ensure the correctness of the skill execution. Owing to the fact that sensors and monitoring algorithms are usually subject to physical restrictions and that effective robot programming is sensitive to the selection of skill parameters, these considerations may lead to different sensor input qualities such as the view coverage of a vision system that determines whether a skill can be successfully deployed in performing a task. Choosing improper skill parameters may cause the monitoring modules to delay or miss the detection of important events such as a mechanical failure. These failures may reduce the throughput in robotic manufacturing and could even cause a destructive system crash. To address above issues, we propose a sensing quality-aware robot programming system that automatically computes the sensing qualities as a function of the robot's environment and uses the information to guide non-expert users to select proper skill parameters in the programming phase. We demonstrate our system framework on a 6DOF robot arm for an object pick-up task., Comment: 7 pages, 9 figures, 1 table; accepted for presentation in IEEE ICRA 2021(IEEE International Conference on Robotics and Automation)

Due to their enhanced mobility and reduced configuration and maintenance cost, wireless control systems (WCSs) are widely used in process and vibration control systems, on medical devices, unmanned vehicles and robotics. However, most literatures in WCSs focus on monitoring and low speed control, and less effort has been made on high speed WCSs. It is because most existing wireless communication protocols cannot provide real-time and reliable communication links with preferable high speed by taking energy saving into consideration.

Due to their enhanced mobility and reduced configuration and maintenance cost, wireless control systems (WCSs) are widely used in process and vibration control systems, on medical devices, unmanned vehicles and robotics. However, most literatures in WCSs focus on monitoring and low speed control, and less effort has been made on high speed WCSs. It is because most existing wireless communication protocols cannot provide real-time and reliable communication links with preferable high speed by taking energy saving into consideration.

The goal of this project is to investigate the fundamental scientific issues that will provide a foundation for a formal design methodology for hard-real-time systems. We think that it is fair to say that the progress made by this group parallels and indeed defines much of the direction and advance in real-time software research in the last 10 years. To bring a scientific discipline to real-time system design, we must develop formal techniques: (1) to specify the real-time behavior of systems (our contributions: RTL, RTCTL, and APTL); (2) to query and validate the desired behavior of a design (our contributions: RTL deductive system, RTCTL model checker, APTL tableau verifier); (3) to ensure that a design can indeed be implemented by proper resource allocation algorithms (our contributions: solutions to adaptive real-time task scheduling, a fast admissibility test for sporadic tasks, complexity results and algorithms for the pinwheel scheduling problem); (4) in addition, we need to demonstrate that our theory can be translated into design tools that form an integral part of a design methodology (our contributions: the Modechart suite of tools). We have obtained a number of basic results in all these areas and also expanded the horizon of real-time computing in the direction of: (a) rule-based systems (our contributions: the EQL, MRL real-time rule-based programming environment); and (b) semantics and concurrency control of real-time data (our contributions: the similarity relation and its use in real-time scheduling). (KAR) p. 4