The rising number of truck-related fatalities on Brazilian highways, alongside global increases in heavy-duty vehicle (HDV) incidents, highlights the need for advanced adaptive safety systems. Responsibility-Sensitive Safety (RSS) provides a verifiable framework for autonomous driving but struggles with HDV challenges such as highly variable mass. This work proposes TruckRSS, an RSS extension for variable-mass vehicles, integrating mass estimation through Neural Networks (NN) and formal runtime verification using Signal Temporal Logic (STL). A data-driven approach employing Gated Recurrent Unit (GRU) models achieved a Mean Absolute Error (MAE) of 121.91 kg after post-processing. Experimental validation in the CARLA simulator demonstrated that TruckRSS adapts braking behavior based on real-time mass estimation, reducing unnecessary following distances and improving traffic flow efficiency compared to standard RSS. Real-time monitoring evaluations confirmed that TruckRSS can be enforced with minimal computational overhead, supporting its scalability for deployment in safety-critical HDV applications.

Ensuring fast, deterministic data exchange is vital for Advanced Driver Assistance System (ADAS), making Time-Sensitive Networking (TSN) a key enabler in real-time automotive environments. ADAS heavily depends on Machine Learning and Computer Vision to analyze information from various sensors, such as LiDAR, radar, and cameras. These technologies support essential functions like adaptive cruise control, collision prevention, and lane-keeping assistance. However, the need to handle and transmit vast amounts of high-resolution sensor data presents substantial computational and networking challenges. This study introduces an predictive model to refine transient subscription management in a publish-subscribe TSN setup for ADAS. The proposed system utilizes predictor to anticipate how long temporary data subscriptions should persist, particularly when triggered by critical scenarios—such as detecting an object at the edge of a sensor’s Field-of-View. By enhancing data exchange efficiency and reliability, this approach bolsters safety-critical CV operations while maintaining the stability of ongoing communication streams. Ultimately, this method improves the resilience and scalability of ADAS in highly dynamic, real-time conditions.

Much effort has been put into Time-Sensitive Networking for Cyber- Physical Systems. Methods and tools have been proposed to support the design and operation of such networks. However, the incorporation of novel technologies, such as computer vision, cloud integration, and decision-making for autonomy, requires an operation regimen that cannot always be modeled at design-time. In this paper, we introduce algorithms to determine whether the load imposed on a Time-Sensitive Networking is within bounds, and to extract formal properties from message headers in order to dynamically verify the network's temporal requirements. We also introduce a formalism based on Signal Temporal Logic to express such properties, a set of Time-Sensitive Networking-specific property monitors, and a lightweight verification engine that can check them at runtime with little overhead. These mechanisms are used by a Safety Enforcement Unit to continuously monitor the network, triggering actions such as best-effort and low-priority traffic cancellations whenever a property fails verification. We evaluated the proposed mechanisms on a prototype autonomous vehicle that uses a Time-Sensitive Networking to interconnect GNSS, IMU, LiDAR, RADAR, and Camera to an ECU. Results demon-strate that these mechanisms can run in real-time, consuming 3% of the processing power of one of the cores in the ECU.

Safety Models for Autonomous Vehicles often neglect fault tolerance, relying on strong assumptions over vehicles’ actuation, such as Responsibility-Sensitive Safety (RSS), which relies on static notions over vehicle’ actuation. This paper proposes to enhance RSS’s proper responses to support fault tolerance during complex maneuvers, specifically overtaking. The proposed approach is carefully built to comply with the original RSS notion of evasive maneuvers. Thus, it can be applied to enable Fault-Tolerant capabilities without losing its original properties. Moreover, the proposed proper responses are modeled using Signal Temporal Logic to promote the verification of system traces using formal methods.

Machine Learning solutions for anomaly detection can be applied in the industry to extend the lifetime of components by promoting actuation in real-time, like optimising machinery parameters to mitigate faults. The development of these approaches usually takes place on Cloud platforms. Nevertheless, Cloud platforms interacting with the real-time control loop add extraneous and unpredictable delays due to service availability and possible communication issues. Thus, embedding anomaly detection solutions in monitoring and control systems is an alternative to enable the aforementioned interaction. In a previous work, we investigated a Deep Autoencoder solution based on vibration data to detect anomalies in Wind Turbines. The solution is currently in usage in a Cloud platform and achieved promising results, which motivated its embedding. This paper demonstrates the process of embedding the solution while maintaining the detection performance metrics. The proposed embedding process reduced the execution time of the preprocessing steps to 18.09% of the original method while saving up to 54.59% of energy on average. An automatic search for low-cost configurations, while evaluating the impact of preprocessing steps, resulted in adequate configurations, improving the original anomaly detection performance and reducing memory utilisation up to 5.5x.

Contemporary Cyber-Physical Systems (CPS), such as autonomous vehicles, are driven mainly by data. Combining timing and data semantics in such Data-Driven systems is crucial to assure safety. This paper proposes an extension of SmartData to support online safety monitoring. By following a Data-Driven Design, we promote a specification of property monitors using Signal Temporal Logic (STL) encompassing Safety Models. Timing aspects from STL specification roots from the timed data intrinsic to SmartData. The property monitors are envisioned as an online monitoring method inside a Safety Enforcement Unit (SEU). The SEU periodically assures the satisfiability of timing and semantics. We demonstrate the proposed design through a case study of an autonomous vehicle modeled using SmartData. The case study considers Mobileye’s Responsibility-Sensitive Safety as a ruler for safety vehicle conditions. Finally, the design provides the online verification capabilities inside the SEU by exploring the interpretation of STL specification as property monitors following the RTAMT library.

In this work, we propose a secure bootstrap protocol for Cyber-Physical Systems (CPS) that compose IIoT Networks. The main contribution of our work is a solution to establish secure communication channels in CPSs through a protocol that enables authentication and confidentiality without the need for constant external verification or pre-stored keys. The proposed protocol relies on the unclonable property of Physical Unclonable Functions (PUF) to build authentication tokens to establish trust between the devices, the gateway, and the Cloud. Devices registration is triggered by an authenticated operator, which informs the PUF responses of the respective device to an External Security Agent (ESA) alongside the identification of the target gateway. ESA and gateway are mutually authenticated using a Certificate Authority and communicate via a secure channel built with HTTPS. The device registration relies on the properties of PUFs to avoid the establishment of security channels via key agreement protocols (e.g., ECDH) and the usage of pre-stored keys. In this way, the PUF challenge response can be used as a secret between the gateway and the device to build trust and establish a secure channel. The presented solution addresses attacks like message replication, Man-in-the-Middle (MITM), and nodes impersonation while supporting gateway integrity check solutions and being free of pre-stored key vulnerabilities.

In this article, we present an Online Learning Artificial Neural Network (ANN) model that is able to predict the performance of tasks in lower frequency levels and safely optimize real-time embedded systems’ power saving operations. The proposed ANN model is supported by feature selection, which provides the most relevant variables to describe shared resource contention in the selected multicore architecture. The variables are used at runtime to produce a performance trace that encompasses sufficient information for the ANN model to predict the impact of a frequency change on the performance of tasks. A migration heuristic encompassing a weighted activity vector is combined with the ANN model to dynamically adjust frequencies and also to trigger task migrations among cores, enabling further optimization by solving resource contentions and balancing the load among cores. The proposed solution achieved energy-savings of 24.97 percent on average when compared to the run-to-end approach, and it did it without compromising the criticality of any single task. The overhead incurred in terms of execution time was 0.1791 percent on average. Each prediction added 15.3585$\mu s$μson average and each retraining cycle triggered at frequency adjustments was never larger than 100$\mu s$μs.

In this paper, we introduce a Framework to Design and Implement Real-time Multicore Schedulers using Machine Learning techniques applied to the very own data such systems produce as they operate. The framework builds on sensors and event counters present in modern hardware platforms and on variables kept by the operating system to capture run-time data that are subsequently subjected to ML tools to produce scheduling heuristics targeting specific optimization goals. It provides non-intrusive mechanisms to collect such data while the system runs real task sets with real workloads, thus preserving the quality of the captured data. It abstracts the Performance Monitoring Unit, thermal sensing, energy monitoring, and Dynamic Voltage and Frequency Scaling available on such platforms through a lean, architecture-independent API. After describing the framework in details, we demonstrate its applicability with the implementation of an energy-efficient, load balancing, real-time, multicore heuristic for a PEDF scheduler. The measured overhead imposed by the framework on the tasks it schedule is at most 0,0003583% and the maximum added jitter is less than 40μs, corroborating the ability of the framework to support the development of effective domain-specific schedulers using machine learning techniques.