Multi-Agent Reinforcement Learning (MARL) - based primary -secondary-control has been shown to exhibit high-performance in Microgrid-based power and energy applications . However, in highly dynamic environments, such as within vehicle-to-grid applications, loss of performance can occur. Specifically, transient loss of accuracy in the synchronization of energy-storage-balance after dynamic topology changes is a known defect, which can overload batteries and reduce stability margins. In this work a newly developed methodology for transient recovery and fault ride-through in battery-based DC-Microgrids is developed and validated. Specifically, an enhancement to MARL-based control utilizing a planned policy with compensation for the DC infrastructure influence is developed, and regional assessment of energy-flow efficiency is examined. The real-time results with quasi-random battery insertion and removal under realistic environmental conditions confirms a reduction in transient recovery time (0.66-13.366%), coupled with enhanced voltage stability (2.637-3.24%) and smoothness (2.9739-3.8462%), better load steadiness (6.666-37.091%), energy saving (2.94%), energy-flow balance enrichment (6.468 %), and raised efficiency (2.626%).

The inherent intractability of probabilistic timing and schedulability analysis of real-time and embedded systems in exact cases has been well discussed in the literature. This has prompted the search for efficient solutions which give meaningful results in practical situations. As many real-time systems are exposed to errors which are random in nature (e.g., due to EMI), it is prudent to consider stochastic schedulability analysis in which the impacts of such errors upon timing are considered. However, work in this area has progressed slowly: in this paper, progress is presented as follows. New schedulability conditions are derived for Earliest Deadline First (EDF)-scheduled periodic and/or sporadic task sets experiencing Binomially distributed random error arrivals, leading to job faults and re-execution attempts. A new analytic model for tightly bounding the impacts of errors over a section of the EDF workload is first presented, and subsequently used to derive an accurate and tractable probabilistic schedulability and timing analysis procedure. The analysis is sufficient to verify that task deadlines are met with a pre-specified upper tail probability R >> 0.5. An example illustrates its application and efficiency. The paper is concluded by highlighting areas of future work.

Demand Response (DR) is seen as one of the key enabling factors in the emerging smart grid. DR takes many forms, including residential smart appliance scheduling. Scheduling algorithms capable of achieving near-minimum cost solutions with low computational overhead are required in order to autonomously respond to varying utility pricing signals. In this paper, the focus is upon an embedded software prototype implementation of a residential load scheduling system. It describes the implementation and testing of a heuristic algorithm for household energy management on a small embedded processor. The performance of the prototype implementation is validated against previously reported experiments and simulations. Test results indicate that the heuristic is efficient enough to be co-located on a small smart meter with limited memory and processing power without any difficulties, helping to open the way for practical consumer demand response.

Industrial wireless sensor networks are often implemented as master-slave type configurations. In this paper, the problem of scheduling transmissions and retransmissions by the master node in order to meet real-time deadlines in the presence of interference is considered. When interference is present retransmissions may be necessary in order to increase reliability of delivery. As packet losses often occur in correlated bursts, it is often useful to insert a gap before attempting a retransmission in the case of failure. In this paper, it is suggested that in many practical cases, the use of a two-stage earliest deadline first scheduler by the master node to assign transmissions and retransmissions to the available slots gives close to optimal performance. Retransmissions are treated as aperiodic requests which are handed by a simple but effective server. An example illustrates the application and effectiveness of the proposed method, and indicates that it performs favorably when compared to two related approaches also based upon deadline scheduling.

The Controller Area Network (CAN) has become a de-facto communication protocol in automation systems over the last three decades. Some CAN networks now employ TDMA-based communication in order to help meet real-time constraints. Whilst this form of media access control brings several timeliness benefits, studies have also illustrated negative effects on transmission reliability; duplicated message instances can help to increase this reliability. In this paper a transmission window technique for CAN is proposed. A bounded amount of re-transmission is allowed for each message within this window, which can in many cases provides increased reliability in the presence of errors or bursts of errors. A probabilistic analysis of transmission windows is presented and used to develop a simple algorithm for calculating the optimal window size to achieve a specified statistical guarantee of message delivery. Stochastic simulations along with computational and empirical results are presented which validate the analysis, and indicate that in many circumstances the technique can potentially reduce the amount of bandwidth needed for specified reliability levels when compared to the use of message duplicates. Suggestions are also made to help increase the reliability of message duplications in error burst environments.

The liberalization of the energy markets, combined with the drive towards a low-carbon economy, has led to the development of the concept of the smart grid. The smart grid is an energy distribution network that enables real-time information exchange for monitoring, control and protection of energy equipment as well as the physical transfer of energy. When smart grid control applications such as generator regulation are implemented over standard IP networks, the stability of the closed-loop has been found to be highly dependent upon the network induced delay, jitter and packet losses. In this paper, we propose and test a fault-tolerant buffering technique and related control design method to enable the effective telecontrol of a generator in such a situation. Computational experiments demonstrate that clear improvements to the closed loop performance and stability can be observed.

Real-time communication networks are often required to operate reliably in harsh environments which expose the system to random errors. Although probabilistic schedulability analysis can employ rich stochastic error models to capture these random behaviors, this is most often at the expense of increased analysis complexity. In this paper, some recent results on probabilistic real-time schedulability analysis are extended to propose an efficient method of time complexity O(n log n) to tightly bound the deadline failure probability for a fault-tolerant real-time network transmitting n periodic/sporadic messages. The paper assumes Earliest Deadline First (EDF) message scheduling is employed and considers both random errors and bursts of errors. A simple example is first used to illustrate the technique, and a more realistic example related to EDF scheduling of a Controller Area Network (CAN) then helps to show the technique has practical value. Although EDF message scheduling is assumed, the technique is readily adaptable to other forms of scheduling.

Real-time computing and communication systems are often required to operate with prespecified levels of reliability in harsh environments, which may lead to the exposure of the system to random errors and random bursts of errors. The classical fault-tolerant schedulability analysis in such cases assumes a pseudo-periodic arrival of errors, and does not effectively capture any underlying randomness or burst characteristics. More modern approaches employ much richer stochastic error models to capture these behaviors, but this is at the expense of greatly increased complexity. In this paper, we develop a quantile-based approach to probabilistic schedulability analysis in a bid to improve efficiency whilst still retaining a rich stochastic error model capturing random errors and random bursts of errors. Our principal contribution is the derivation of a simple closed-form expression that tightly bounds the number of errors that a system must be able to tolerate at any time subsequent to its critical instant in order to achieve a specified level of reliability. We apply this technique to develop an efficient 'one-shot' schedulability analysis for a simple fault-tolerant EDF scheduler. The paper concludes that the proposed method is capable of giving efficient probabilistic scheduling guarantees, and may easily be coupled with more representative higher-level job failure models, giving rise to efficient analysis procedures for safety-critical fault-tolerant real-time systems.

It has been widely accepted that real-time implementations of feedback control systems can be susceptible to timing jitters which may be caused by the underlying real-time and/or embedded implementation architecture. Previous experimental studies aimed at quantifying the levels of degradation that may be observed have mainly concentrated upon relatively simple fixed-gain feedback control schemes (e.g. PID) and time-invariant plant. Although some degradation has been observed, most systems have been shown to be surprisingly robust unless driven to extreme limits. In this paper, we study the jitter sensitivity of a real-time embedded implementation of a digital parameter-adaptive control system. The purpose of the study was two-fold; primarily, to obtain empirical data related to jitter sensitivity (as measured by a quadratic performance metric), and secondarily to explore the potential impact of the underlying scheduler choice on system behavior. The findings indicate that the adaptive controller was heavily influenced by sampling jitter, and that the choice of task scheduler had a role to play.

It is well known that the performance of real-time control applications can be seriously degraded by sampling and actuation jitters and delays. This paper describes a simple hybrid-EDF algorithm which may help to ameliorate some of these problems in resource-constrained embedded systems in which the use of simplified and non-preemptive schedulers is required. The algorithm supports a single preemptive time-critical sampling and actuation task, whilst the remaining tasks are executed non-preemptively and have implicit deadlines. The paper develops an efficient schedulability analysis technique for this framework; analysis and an example are given to illustrate the improvement in efficiency over related techniques.

It is well known that the performance of real-time control applications can be affected by many factors, including the choice (and configuration) of the hardware platform and task scheduling algorithm. This paper presents a real-time hardware-in-the-loop simulation tool that simultaneously integrates concepts from control and real-time computing to allow further investigation into these links. It allows the experimental prototyping and analysis of real-time control systems on real hardware - but within a simulated plant environment -and thus allows the potential impacts of hardware and software architecture on control performance and dependability to be evaluated. The structure of the proposed tool is described, along with a preliminary experiment that has been performed to illustrate some of the test facility capabilities.

Highly dependable distributed embedded systems (DES) have traditionally been developed using static approaches, i.e., assuming a mostly constant environment. However, the little flexibility of such approaches does not allow continuous operation under dynamic environments. The Flexible Time-Triggered (FTT) communication paradigm is a promising approach to introduce the required flexibility. However, for continuous operation reliability is also crucial. Replicated star topologies are particularly well-suited to provide an increased reliability. Nevertheless, for FTT-CAN, the implementation of FTT for CAN, no replicated star topology that takes advantage of FTT-CAN's features to increase reliability and error containment exists. This paper discusses important design questions that need to be solved to create such a novel solution.

The potential benefits that adaptive and predictive control schemes can bring to building environment control applications such as HVAC have been well documented in recent years. One significant drawback of these schemes is that in many cases they require large computational burdens, especially in adaptive situations and when constraints are present. A secondary drawback is the large number of parameters that have to be potentially tuned. As high-bandwidth real-time embedded computing platforms and advanced levels of control knowledge may not be realistic expectations for many users, the applicability of advanced schemes is currently limited for most Smart home applications. In this paper, a prototype infinite-horizon `plug-and-play' adaptive MPC scheme is presented for input-constrained HVAC applications. The scheme has been optimized for real-time implementation on small, low-cost embedded processors and requires little user preconfiguration. Preliminary hardware-in-the-loop based experimental results indicate good performance of the technique coupled with extremely low overheads.

The use of wireless communications in real-time control applications poses several severe problems related to the comparatively low reliability of the communication channels. This paper is concerned with application-level strategies for ameliorating the effects of packet losses and burst errors in sampled-data control systems implemented via one or more wireless links. In particular, the paper develops an adaptive/predictive compensator that reconstructs the best estimates (in a least squares sense) of a sequence of one or more missing process data packets. An embedded implementation of the proposed technique is applied to a case study, and it is shown that the proposed techniques outperform existing methods in a control system experiencing artificially induced burst errors, whilst also exhibiting acceptably low overheads.

Limited-preemption forms of uniprocessor scheduling provide a practical trade-off between flexibility and system overheads in embedded kernels. This paper considers the limited-preemption scheduling of implicit deadline tasks using the EDF algorithm. Upper bounds for the occurrence of a deadline miss in these task sets are derived, and employed to create several sufficient schedulability tests of increasing complexity and tightness. Exact analysis is then considered, and an algorithm with polynomial-time complexity is developed for situations in which the CPU utilization is bounded to be less than unity. Experimentally, the sufficient tests are shown to have acceptance ratios increasing from ≈60% up to ≈ 99%, and the number of evaluated deadlines required for the exact test is reduced by several orders of magnitude over previously known techniques.

Many distributed control systems employ TDMA-based communication over CAN in order to meet realtime constraints. Whilst this form of media access control brings several timeliness benefits, studies have also illustrated negative effects on transmission reliability. This paper extends the ‘window transmission’ technique which was recently proposed by the authors to help overcome this problem in TDMA-based networks to include the effects of correlated (burst) errors. This paper employs a simple Markov model to describe burst error behaviors in a CAN network, and the model is used to develop an algorithm for calculating TDMA slot sizes which aim to provide prespecified statistical guarantees of message delivery. Computational results are presented which indicate that the technique can reduce the amount of bandwidth needed for specified reliability levels by a significant factor when compared to the use of message duplicates. The paper is concluded with an empirical study which provides further supportive evidence for the described technique.

The management of tasks is an essential requirement in most real-time and embedded systems, but invariably leads to unwanted CPU overheads. This paper is concerned with task management in real-time and embedded systems employing the Earliest Deadline First (EDF) scheduling algorithm. Currently, the best known techniques to manage EDF scheduling lead to overheads with complexity O(log n), where n is the number of recurring (periodic/sporadic) tasks. In this paper it will be shown that if both the ready and waiting queues are represented by either i) timing and indexed deadline wheels or ii) digital search trees, then all scheduling decisions may be made in time proportional to the logarithm of the largest time representation required by the system, pm. In cases where pmis relatively small, for example in some embedded systems, extremely efficient task management may then be achieved. Experimental results are then presented, and it is shown that on an ARM7 microcontroller, when the number of tasks is comparatively large for such a platform (> 250), the worst-case scheduling overheads remain effectively constant and below 20 ¿s. The results indicate that the techniques provide some improved performance over previous methods, and also seem to indicate that there is little discernable difference between the overheads incurred between employing a fixed- or dynamic-priority scheduler in a given system.

Embedded control systems play an increasing role in many safety critical system designs. The correct and dependable implementation of such systems depends on many factors, including the design of system hardware, software and fault tolerance mechanisms, the choice of programming language, and also the testing, verification and validation techniques employed. In this paper, a set of guidelines for the development of dependable embedded systems is presented. Although the paper is primarily concerned with single-processor applications, extensions to multiprocessor systems are discussed where appropriate. Although the creation of dependable embedded systems cannot simply rely on the enforcement of several such rules or guidelines, experience gained from several years' experience of teaching, research and development in these areas indicates that adherence to a small, but workable, set of rules and guidelines can avoid many of the traps and pitfalls commonly encountered in the creation of dependable embedded systems.

In this paper, we explore the impact of task preemption on the dependability of a single-processor embedded control system. Our particular focus in this exploratory study is on static-priority, time-triggered scheduler architectures. The study is empirical in nature and we employ a hardware-in-the-loop (HIL) testbed, representing a cruise control system for a passenger vehicle, in conjunction with fault-injection to perform the dependability comparisons. The results we have obtained suggest that the presence of preemption may have a negative influence on dependability; however further work is needed in this area before more general conclusions may be drawn.

The controller area network (CAN) protocol was originally introduced for automotive applications but is now also widely used in process control and many other industrial areas. In this paper, we present a low-cost redundancy-management scheme for replicated CAN channels that helps to ensure that clocks (and, hence, tasks) on the distributed nodes remain synchronized in the event of failures in the underlying communication channels, without the need for expensive or proprietary interface electronics. We argue that, when using this framework with duplicated channels, the probability of inconsistent message delivery drops to acceptable levels for a wide range of systems. Through an analysis of the protocol and a case study, we conclude that the creation of reliable, low-cost, distributed embedded systems using CAN is a practical possibility.