Ing. Pavel Píša, Ph.D.

The article informs community and companies about CAN bus related projects and results achieved and offered by the Czech Technical University, Faculty of Electrical Engineering. The article informs regarding CAN FD controller open-source IP core developed at Czech Technical University and current integration and release of the corresponding driver in mainline Linux operating system kernel. Emulation of the CTU CAN FD IP core is available in mainline QEMU emulator as the result of previous work. The use of own CAN controller boards and ecosystem for testers delivered to Skoda Auto and prototype of precise CAN (FD) latencies evaluation are described as well.

The summing integration (SumInt) analog to digital conversion (ADC) technique combines and preserve many of excellent features of double integration and sigma-delta ADCs. It is well suited for application where integral of continuously sensed input signal carries information to acquire. It has been initially conceived at the PiKRON company for digitizing compounds responses measured by UV-VIS spectrophotometric detectors in high-performance liquid chromatography systems. The compound concentration in the sample is proportional to the integral/area under response peak. In a contrast to double integration ADC, the SumInt ADC integrates input signal continuously and does not require reset/idle interval control. In comparison to sigma-delta ADC, the frequency of reference switching is much lower (less charge leakage). The paid price is sampling interval floating in a range of up to one half of the fixed modulator interval. The actual ESA funded De-Risk project focuses on reuse of the technique for low analog components count conversion in radiation tolerant systems where FPGAs are already in use.

Abstract—This paper presents a RISC-V based computer system simulator, QtRVSim (https://github.com/cvut/qtrvsim), designed for teaching and learning computer systems principles. The simulator allows students to run assembly programs and observe the instruction execution on single-cycle and pipelined microarchitectures. The simulator graphically displays the major components in the datapath, including the register file, the arithmetic-logic unit, memory caches, peripherals, and the control unit with control signals. QtRVSim is free and open-source software available on GitHub and as a WebAssembly application online. For additional materials and related computer architecture projects, see https://comparch.edu.cvut.cz/.

This paper deals with the explicit model predictive control (MPC) algorithms for permanent magnet synchronous motors (PMSM). The algorithms generate continuous and smooth set of pre-computed control laws represented by parameterized gains. The selection and application of the gains in real motion control of PMSM is explained. The MPC design introduces cost functions and control laws that define explicit algorithms. For this purpose, a unified state-space model of PMSM is proposed. In the paper, practical aspects of considered explicit MPC algorithms are investigated for control systems with floating-point arithmetic and usual incremental position sensors with quantized values. The proposed solution is demonstrated by real experiments with 95W PMSM controlled by Field-Programmable Gate Array (FPGA) unit.

V tomto článku bude popsané, jak je možné jádro Linux upravit tak, aby byla reakční rychlost použitelná pro řešení řídicích aplikací. V dalším návazném článku pak bude demostrováno, jak zkonfigurovat a sestavit jednoduchý zpětnovazební systém s destičkou Raspberry Pi pro řízení stejnosměrného motoru s minimálním počtem přidaných komponent. Hardware je vybraný tak, aby byl snadno dostupný, levný a celý návrh je koncipovaný tak, aby byl vhodný a dostatečný pro výuku. Pokud bude zájem, na dvojici článků mohou navázat další, již zaměřené na řešení přibližující se postupně návrhu systémů použitelných v reálním nasazení.

Článek navazuje na předchozí úvodní článek, ve kterém bylo popsané, jakým způsobem byly využity konstrukce spinlocků pro dosažení možnosti téměř okamžitého přeplánování úlohy na jinou s vyšší prioritou i během provádění operací/systémových volání uvnitř jádra. Dlouhodobé sledování dob odezvy systému i při maximální zátěži ukazuje, že pro výkonnější systémy na bázi architektury x86 nepřekročí maximální doby odezvy okolo 100 μs, pro nevýkonné systémy jako je Raspberry Pi 1 maximální doby odezvy nepřekročí 350 až 400 μs.

The paper deals with a description of principles and application concepts of a distributed industrial manipulation system containing wireless operated moving manipulation units. In the paper, there is a novel application of the wireless data communication. The solution is intended for industrial manipulation systems supporting robotic plants and centres. An example of such systems used in this paper consists of several moving manipulation units, several stationary auxiliary units and control computer.

Fast, easy and frequent/continuous testing is critical for embedded operating systems and applications development. The QEMU system emulator provides a solution to easily test developed operating systems kernels and drivers on multiple emulated architectures and board models. But emulation of automotive and industrial control buses is not available in QEMU yet. Presented project implements CAN controller emulation for QEMU. The emulated controllers can be connected to another ones inside the QEMU instance or can be connected to the host system virtual or physical CAN bus network if GNU/Linux is used as QEMU host system. The PCI based CAN bus interface card was selected as the first target for our work because it can be easily attached to all QEMU architectures that support PCI/PCIe.

Linux kernel contains a full featured CAN bus networking subsystem. It can be accessed from applications via several different interfaces. This paper compares the performance of those interfaces and tries to answer the question, which interface is most suitable for capturing traffic from a big number of CAN buses. Motivation for this work is the development of various CAN traffic analyzers and intrusion detection systems. Besides traditional UNIX interfaces we also investigate the applicability of recently introduced 'low-latency sockets' to the Linux CAN sub-system. Although the overhead of Linux in general is quite large, some interfaces offer significantly better performance than others.

Matlab/Simulink is a commercial tool used by many engineers and researchers worldwide to design and develop various systems, usually containing a lot of mathematical computations. Initially, Simulink was intended for performing simulations of dynamic systems (hence the name), but nowadays it also allows to create their prototypes or even final implementations. The system (for example a motor controller) is first designed in a graphical way in the form of a data-flow graph and then the Embedded Coder tool (a part of Simulink) is used to generate the C code directly from the graphical model. Simulink Embedded Coder already contains support for several popular embedded boards running Linux such as Raspberry Pi or BeagleBoard but for unknown reason, this support can only be installed on Windows hosts. Moreover, the code generated for these targets has problems with precise timing when run on Linux with real-time (preempt_rt) patches. In this paper we describe a custom developed Embedded Coder target ert_linux that does not suffer from the above mentioned shortcomings and is freely available for use. We also describe a few applications developed with this target, for example a simple motor controller with the Raspberry Pi that can be used for education. The other applications show that ert_linux-based solutions can be used even for more demanding applications with sampling frequencies around 20 kHz.

Real-Time Publish-Subscribe (RTPS) is a protocol, based on UDP/IP,that allows easy and efficient implementation of data-driven distributed real-time applications. The protocol was adopted as an OMG specification and it is intended as an interoperability protocol for applications based on the Data Distribution Service (DDS) API. RTPS is widely used in many industrial applications and it has both commercial and open source implementations. One open-source implementation is called Open Real-Time Ethernet (ORTE) and it is known to work on many platforms including GNU/Linux, Windows,FreeBSD and MacOS. In this paper we describe a new addition to the supported platforms, which is an Android operating system for mobile devices. We provide an overview of the steps that were needed to port the protocol to the Android platform. We ported the existing native library and added a Java wrapper around it. We comment on the troubles we had and their solutions. We compare the performance of ORTE on Android with other platforms and we present our Android phone application for controlling and monitoring a mobile robot.

The LIN-bus (Local Interconnect Network) is a vehicle bus standard or computer networking bussystem used within current automotive network architectures to control slave peripherals for which CAN bus is too expensive or complex. Concept of LIN frames and identifiers has its roots in CAN-bus however data bytes are serialized in asynchronous serial communication format usual for UARTs. The article focuses on LIN-bus support implementation and integration into Linux based systems with attempt to offer portable solution with minimal hardware dependencies. This is possible thanks to the compatibility of common UART hardware with LIN-bus serial format. The developed slLIN protocol driver is implemented as a Linux TTY line discipline and uses only common Linux UART serial line discipline API. The solution does not require to implement specialized driver for each architecture or serial interface hardware and is highly portable. The interface from application to slLIN is based on CAN protocol family network API (same as SocketCAN uses). This approach was chosen because LIN-bus is usually found together with CAN-bus infrastructure in applications. The portability of the implementation has been tested on common PC serial port and MPC5200 hardware against third party fully-functional LIN device. A utility for frame sequences configuration is also part of the implementation.

The paper deals with a novel application of the wireless data communication for a continual control of the distributed manipulation systems. The solution is intended for industrial robotic plants. A considered way of the wireless communication is based on ZigBee protocol. The protocol is tested for a real time bidirectional data communication within a manipulation system. The system consists of several moving manipulation units, several stationary auxiliary units and one control computer. The computer provides the cooperation of all units in the system in relation to the user requirements. The system is controlled by a simple feedback multi-level control realized in MATLAB - Simulink environment. The paper is focused on the realization of the boards of power electronics, transmitters, optical positional sensors, optical gates and their networking in accordance with ZigBee protocol definition. The behavior of the ZigBee is illustrated by several records from real experiments.

The article describes implementation of UIO and Comedi drivers for Humusoft MF624 and MF614 data acquisition cards. Basic functions (D/A, A/D converters, digital inputs/outputs) of Humusoft MF624 card were implemented into the Qemu emulator as well which enable to experiment with drivers implementation without physical access to the cards and risk of data lost when drivers are developed and tested on same primary Linux kernel instance. The article can help newcomers in the area to gain knowledge required to implement support for other similar cards and hardware emulation of these cards. The matching real and virtual setup can be used in operating system courses for practical introduction to simple drivers implementation and helps with understanding internal computation world with real world computers interfacing.

The paper deals with a realization of control of distributed mechatronic system operated from MATLAB-Simulink environment. The mechatronic system serves as a laboratory model of simple manipulation system. The communication within the system is provided by wireless ZigBee protocol through a coordination unit. This unit is connected via serial interface to the control computer, where appropriate Simulink model runs. The paper addresses briefly individual physical components of the system (units, passive and active or executive elements etc.), specific blocks of used Simulink model and demonstrates the system behavior on several examples.

The aim of this paper is to describe the method used to analyze timing properties of a Linux-based CAN-to-CAN gateway and to briefly present some of the analysis results. The considered gateway, implemented as an embedded system, can significantly simplify rapid prototyping of CAN-based distributed systems such as those used in automotive applications. The main interest was to measure the latencies induced by the gateway under the wide range of conditions such as various types of traffic, gateway configurations, additional system load and Linux kernel versions. The combinations of these conditions resulted in almost 400 different experiments and more than 600 graphs visualizing the results. A simple method to automate the experiments and to effectively draw conclusions from such a high number of experiments is proposed.

In this paper, we present a middleware for providing distributed soft real-time applications with a uniform API for reserving heterogeneous resources with real-time scheduling capabilities in a distributed environment. The architecture relies on standard POSIX OS facilities, such as time management and standard TCP/IP networking services, and it is designed around CORBA, in order to facilitate modularity, flexibility and portability of the applications using it. However, real-time scheduling is supported by proper extensions at the kernel-level, plugged within the framework by means of dedicated resource managers. Our current implementation on Linux supports the reservation of the CPU, disk and network bandwidth. However, additional resource managers supporting alternative real-time schedulers for these resources, as well as additional types of resources, may be easily added.

The uLan protocol is the multi-master communication protocol aimed on small RS-485 control networks. It provides deterministic media access arbitration and it is open in design from its origin. An open-source implementation of the protocol has already been available for many years. The article focuses on its adaptation for use in distributed home appliances (switches, lights and HVAC components interconnection and control). Because devices do not have resources to mutually examine their often large objects/properties dictionaries, the mechanism to map properties values into process data messages slots has been implemented. The message slots act as (virtual) wires which are setup by configuration tools running on PC. Examples of developed devices using developed concept are presented at the end of the article together with tools available to help with fast prototyping of new devices and their testing in PC environment.

In this paper, we thoroughly analyze timing properties of CAN-to-CAN gateway built with Linux kernel CAN subsystem. The latencies induced by this gateway are evaluated under many combinations of conditions, such as when traffic filtering is used, when the gateway is configured to modify the routed frames, when various types of load are imposed on the gateway or when the gateway is run on different kernels (both rt-preempt and vanilla are included). From the detailed results, we derive the general characteristics of the gateway. Some of the results apply not only for the special case of CAN-to-CAN routing, but also for the whole Linux networking subsystem because many mechanisms in the Linux networking stack are shared by all protocols. The overall conclusion of our analysis is that the gateway is in pretty good shape and our results were used to support merging the gateway into Linux mainline.

The aim of this paper is to introduce LinCAN, a CAN driver system for Linux, developed at the Department of Control Engineering of the Czech Technical University in Prague, and to provide a thorough comparison with SocketCAN, which is the most common CAN solution for Linux nowadays. Thorough timing analysis and performance comparison with Socket CAN are presented, with several case-studies and applications of LinCAN shown in the end. A timing analysis and performance tests of both drivers were performed using various types of load with several recent Linux kernels. Obtained results indicate that LinCAN seems better suited for hard real-time applications, its performance being either better or on-par with SocketCAN in presented tests.

CAN-bus (Controller-Area Network) is widely used communication bus in industrial and automotive systems. Besides that it simplifies wirings, CAN-bus offers deterministic medium access algorithm for which it is very suitable for certain type of time-critical applications. Typical CAN-based network consists of multiple microcontrollers and one or a few more complex nodes which implement higher-level control algorithms. These more complex nodes can run Linux OS so there is a need for CAN-bus drivers. There exist several projects which offer CAN drivers for Linux. Probably the most known driver is Socket-CAN, which is already partially included in the mainline kernel. Another option is LinCAN driver. The basic difference between these two drivers is that Socket-CAN is built around standard Linux networking infrastructure, whereas LinCAN is a character device driver with its own queuing infrastructure.

Parallel robots represent way to considerably improve accuracy and speed of industrial machine tools and their centres. This paper deals with the preparatory operations: homing and calibration, which precede the start-up of the robot work, i.e. real control process. Their procedures are discussed with respect to planar parallel robots and their control. In this paper, as a suitable control strategy, the model-based predictive control is considered. The predictive control offers operator to continuously influence the control process. The control issues relating to planar parallel robots are discussed here.

The paper deals with the concepts of model-based control and trajectory planning intended for industrial parallel robots. These robots are characterized by very good dynamical properties arisen from small number of moving masses in comparison with conventional configurations. In the paper, multi-level (hierarchical) control will be investigated. It can be specified as a model-based control providing positional and speed loops with addition of fast low-level current-loop control. As a suitable representative of model-based control, predictive control is considered. Described concept can offer more possibilities to manage the control process than usual cascade control. Finally, the paper outlines two different concepts of trajectory planning. The first concept considers only pure geometrical features (curve-based planning) without relation to the real robots. The second concept conversely takes into account the dynamical features of the real robot with initial and final points.

This contribution deals with optimal reconfiguration of FPGA modelled by positive and negative time lags and changeover times. Particular focus is on posssible inclusion in the framework for contracts in embedded systems.

The paper deals with the design-by-optimization and control of redundantly actuated parallel kinematics sliding star.

Tento článek popisuje univerzální modul pro systém PC/104. Modul je určený pro aplikace řízení pohybu současně až čtyř pohonů vybavených IRC snímači. Jádrem jednotky je programovatelné logické pole, které může také zajišťovat obsluhu rozhraní sběrnice CAN a PROFIBUS pro připojení dalších zařízení k modulu. V programovatelném poli může být implementováno procesorové jádro a tak může celý modul pracovat samostatně bez základní desky PC/104. Navržené řešení umožňuje změnu funkcí programovatelného pole za běhu z nadřazeného počítače PC/104, což zvyšuje flexibilitu modulu.

The paper presents a new analog signal to digital representation conversion technique. The proposed approach tries to combine and preserve significant excellent features of double integration and sigma-delta analog to digital converters. The proposed AD-converter is constructed for a direct conversion of two analog signals ratio to the digital domain. The design uses an oversampling technique and the digital output resolution and bandwidth can be tuned by digital data post-processing. Other outstanding feature is an accurate area under signal curve measurement. The output data rate is constant (unlike for voltagefrequency converters) and the response time is maintained short without need of high modulator switching frequency. The described AD-converter features are a big advantage for the spectrophotometric absorbance measurement.

The uLan protocol is message oriented multi-master communication protocol for microcontrollers. It uses 9-bit character format for addressing and link control and RS-485 physical layer. The protocol is used in HPLC instruments and analyzers for instruments control and data acquisition. The protocol has potential to be used in the home automation applications and some other industrial field applications for its open nature design and proven implementation. The software support is provided for more RS-485 cards, converters for PC RS-232 ports and embedded microcontrollers. One of applications utilizing uLan communication is CHROMuLAN HPLC system software, which provides affordable solution for HPLC assay mathematical data processing and complex HPLC instruments sets control over uLan communication network. Other successful field example utilizing uLan communication is agriculture automatic feeding system.

Predictive control is one of control strategies, which optimize control actions within certain time range. Control actions are not designed only for current state of the system, but also they are optimized for the following time instants. Presented paper makes simple overview of Predictive control algorithms and their tests under MATLAB.SIMULINK environment. Furthermore, it presents Web publication of the MATLAB scripts and functions and SIMULINK schemes used for Predictive control.

The paper describes structured model based control of redundant parallel robot kinematics.

The paper describes the concept and implementation of two advanced approaches towards model based control of redundant parallel kinematical machines. It is general predictive control and sliding mode control together with traditional PID control.

This article deals with a distributed real-time application modelling by timed automata. The application under consideration consists of several processors communicating via Controller Area Network (CAN); each processor executes an application that consists of tasks running under an operating system (e.g. OSEK) and using inter-task synchronization primitives. For such system, model checking algorithm implemented in a model checking tool (e.g. UPAALL) can be used to verify complex time and logical properties of the proposed model (e.g. end-to-end response time, state reachability, deadlock freeness). Since the proposed timed automata model contains more crucial details of the system behavior with respect to classical approaches to the response time analysis, the model checking approach provides less pessimistic results in many cases.

The deliverable deals with the project OCERA. The main objective of the project is the creation of a library of free-software components for the design and development of real-time and embedded systems (RTES).