DISTRIBUTED CONTROL SYSTEMS BASED ON COTS

COMMUNICATION DATA BUS

Martin S

ve´da, Istva´n Szabo´ and Vladimı´r Oplusˇtil

Unis spol. s r.o.

Jundrovska´ 33, Brno, Czech Republic

Keywords:

Distributed control system, COTS data bus, CAN, CANaerospace, mobile robot, avionics.

Abstract:

This paper deals with the distributed commercial off the shelf (COTS) data bus based on Controller Area

Network (CAN) used as a communication data bus for Autonomous Locomotion Robot (ALR) and a System of

Avionics Modules (SAM) that is used in civil aircraft Ae270. This article describes main characteristics of CAN

communication data bus and its higher layer protocol CANaerospace that are used in communication system

of ALR and SAM. The basic idea of distributed control systems are described and their main characteristics are

presented. Developed control systems proved that the CAN with HLP CANaerospace is efﬁcient and reliable

communication data bus that can be used in safety critical applications like mobile robots, automotive, and

avionics systems.

1 INTRODUCTION

The Controller Area Network (CAN) adapted by

Bosch is known as a protocol for high performance

and high reliable serial communication links between

electronic control units in the ﬁeld of automotive and

industrial control applications. Design of CAN data

bus enables distributed system control in real-time

with a high degree of transmission security.

Since that prime manufacturers of integrated cir-

cuits already implement CAN protocol support into

their product, rapid exploitation of this protocol with

other applications, e.g. robotics and aviation will take

place.

Distributed control, or distributed problem solving,

involves the use of decentralized, loosely coupled con-

trollers or problem solvers. The system is decentral-

ized, so that both the control and the data are func-

tionally and often geographically distributed.

Multiprocessor control systems that are charac-

terised with high independence comprise very impor-

tant area of automation resources. The spine of these

systems must include a powerful communication sys-

tem. Maintenance of the multiprocessor system pri-

ority requires a fast, safe, and reliable communication

channel. Based on these requirements, the CAN com-

munication device was used (CAN).

2 CAN COMMUNICATION

SYSTEM

The distributed control system based on CAN com-

munication data bus allows nondeterministic control

in real-time. Matching rate and access method of

CAN data bus is speciﬁcally adapted for distributed

control systems. CAN is based on broadcast com-

munication mechanism which is achieved by using

a message oriented transmission protocol. Stations

and their addresses are not deﬁned, CAN only deﬁnes

messages. The message is identiﬁed by a message

identiﬁer which is unique within the whole network

and it deﬁnes not only the content but also the priority

of message.

2.1 Higher Layer Protocol

CANaerospace

CANAerospace (see (CANaerospace)) is an ex-

tremely lightweight protocol/data format deﬁnition,

which was designed for the highly reliable commu-

nication of microcomputer-based systems in airborne

applications via CAN. The purpose of this deﬁnition

is to create a standard for applications requiring an

efﬁcient data ﬂow monitoring and easy time-frame

synchronisation within redundant systems. The deﬁ-

Švéda M., Szabó I. and Opluštil V. (2006).

DISTRIBUTED CONTROL SYSTEMS BASED ON COTS COMMUNICATION DATA BUS.

In Proceedings of the Third International Conference on Informatics in Control, Automation and Robotics, pages 82-85

DOI: 10.5220/0001215300820085

 SciTePress

nition is kept widely open to allow implementation of

user-deﬁned message types and protocols.

The data format deﬁnition speciﬁes 6 basic mes-

sage types, which are used for different network ser-

vices. Each message type has an associated CAN-ID

range deﬁning the message priority. The identiﬁer as-

signment within the speciﬁed ranges is at the user’s

discretion.

Canaerospace features:

• unique identiﬁcation of message format by mes-

sage transmitting device identiﬁer and message data

type,

• message numbering for identiﬁcation of message

drop-outs,

• critical event signaling to enable each bus-linked

device to inform others of its error status

• assigning addresses to bus-connected devices and

service priority,

• ﬁxed assignment of identiﬁers to default data values

used in aerospace technology, mechanism similar to

that applied to ARINC429 bus,

• easy and unique implementation of CANAerospace

protocol simpliﬁes certiﬁcation process,

• high reliability.

3 DISTRIBUTED CONTROL

SYSTEM FOR ALR

In practice, a multiprocessor distributed system con-

sists of several independent processor modules. This

structure of control system has been successfully im-

plemented and tried out on the project of autonomous

locomotive robot VUTBOT2 (Szabo´, 2002), (Szabo´

et al., 2002), (Szabo´, 2003).

Conception of distributed multiprocessor system al-

lows to easily connect next modules into the system

without spurious interference to existing hardware and

software parts.

The control system consists of three main, mutually

relating parts:

1. Multiprocessor control system of undercarriage.

The ALR undercarriage control system is divided

into several function blocks.

2. ALR monitoring module. This part is used for

monitoring internal and external ALR states.

3. Higher layer control module. This system repre-

sents input/output communication gate and can be

used for speciﬁcation and monitoring of ALR ac-

tivities by a master control system.

Figure 1: ALR control system.

Multiprocessor control system of undercarriage is

divided into several functional blocks:

• Communication module. Module contains radio

FM transceiver for wireless communication with

manufacturing system.

• Sensor module. An external sensorial system and

a laser distance measurement scanner PLS are con-

nected to this module. (PLS device presents the

main external sensor in movement control on the

working environment).

• Locomotion module. This module supports drive

and regulation of DC motors. Front wheel speed

difference is calculated (electric differential) based

on the required direction. The module allows man-

ual control of motion in a critical situation.

• Control module. The module allows robot motion

control according to basic commands describing

optimal robot path to the goal position. This module

analyses and performs these basic commands that

can be extended upon additional demands.

The multiprocessor control system is con-

nected by CAN (CAN) communication

data bus with higher layer protocol (HLP)

CANaerospace (CANaerospace).

4 DISTRIBUTED CONTROL

SYSTEM FOR SAM

The SAM – System of Aviation Modules is a unique

open distributed aviation system that performs spe-

ciﬁc functions onboard the Ae270 aircraft (see Fig. 2)

and also provides internal functions for module di-

agnostics. Each module integrates and automatically

carries out speciﬁc functions onboard the plane thus

increasing the comfort of the cockpit crew when oper-

ating and controlling the aircraft. The individual mod-

ules can operate independently. Alternatively they can

DISTRIBUTED CONTROL SYSTEMS BASED ON COTS COMMUNICATION DATA BUS

be extended or grouped as required by the customer

via a communication bus implemented by means of

the CAN (Controller Area Network) serial commu-

nications interface standard and the HLP (Higher

Layer Protocol) communication protocol within the

CANaerospace standard.

Figure 2: SAM modules onboard the Ae270 aircraft.

The hardware and software resources have been

developed using elementary function blocks shared in

general by the individual SAM modules. Universality

and modulability of SAM system insures compatibil-

ity with other civil airborne applications.

Short description of each module in SAM is given

in Table 1.

Table 1: Description of individual SAM modules.

Module Functional description

SAM IUCH Electric power supply control

module

SAM TIM Timer, fuel distribution and de-

icing module

SAM WSHD Windshield heating module

SAM ACC Y–load factors monitoring mod-

ule

SAM INP Module of inputs

SAM FUEL Fuel distribution control module

SAM ENG Engine monitoring module

SAM MFD Multifunctional display module

The SAM was primary developed for use in the

Ae270 aircraft, but it can be used in any other avion-

ics system. The SAM was developed in accordance

with RTCA/DO-254 and RTCA/DO-178B standards.

Compliance of these standards during the develop-

ment cycle is required by a certiﬁcation authorities for

a type certiﬁcation issue.

4.1 Evaluation of CAN Data Bus in

Ae270 Civil Aircraft

Implementation of SAM distributed control system

based on CAN communication data bus into the Ae270

civil aircraft was one of the ﬁrst similar projects in the

world. Therefore certiﬁcation authority requires many

simulations and tests to be performed to verify usabil-

ity and reliability of CAN data bus as communication

bus for control system used in civil aircraft.

4.1.1 Bus Load and Bus Response

For bus load and bus response evaluation the mathe-

matical models were designed. The detailed formula-

tion of these models can be seen in (CRI2).

Bus load evaluation was performed for three basic

screens of the SAM MFD module. These screens are

labeled HOME1, HOME2, and HOME3 and results

are shown in Figure 3.

11.5

12.5

13.5

14.5

15.5

0 10 20 30 40 50 60

Bus Load [%]

Time [s]

Home1

Home2

Home3

Figure 3: Bus load for three basic screens of SAM MFD

module.

Mathematical model was veriﬁed by measurement

during real operation using CANalyzer (CANalyzer)

and measured results match theoretical results well.

From presented data results that average bus load is

less than 15 % even during the occurrence of error

messages on the bus, thus bus capacity for SAM is

more than sufﬁcient.

Bus response means time that pass from data inser-

tion into the queue to their receiving by a receiver.

The results obtained from mathematical model are

shown in Figure 4. The ﬁgure describes relation be-

tween CAN identiﬁer, which designates the message

priority transferred on CAN bus, and calculated time

response.

From the given ﬁndings results effect of message

priority identiﬁer on prolongation of bus response, but

ICINCO 2006 - INTELLIGENT CONTROL SYSTEMS AND OPTIMIZATION

0.05

0.1

0.15

0.2

0.25

0 50 100 150 200 250

Bus Response [s]

CAN ID x 10e-1

Home1

Home2

Home3

Figure 4: CAN bus response.

during normal operation of SAM modules the maxi-

mum theoretical response time can not be achieved.

4.1.2 Bus Characteristics

The basic characteristics of CAN data bus used on-

board the Ae270 aircraft are:

• transfer rate 27.7 Kbps,

• average load of CAN bus on Ae270 is 33.9 %,

• probability of undetected message corruption in a

CAN network is around 1 · 10

−13

• probability of undetected failures per ﬂight hours is

2.7 · 10

−8

• length of CAN data bus is 13.5 m,

• the number of nodes (SAM units) is 8.

System of Avionic Modules SAM was awarded the

Gold Medal of 43th International Engineering Fair

in Brno, Czech Republic, 2001. In December 2005

the Unis company obtained an approval from Czech

aviation authority (CAA) for its System of avionics

modules (SAM). Next in December 2005 the Euro-

pean Aviation Safety Agency (EASA) issued a type

certiﬁcate for an Ae270 civil aircraft produced by Aero

Vodochody. Speciﬁc functions onboard the Ae270 air-

craft is performed by SAM developed by Unis. The

type certiﬁcate for US market was issued by the Fed-

eral Aviation Agency on February 24, 2006.

5 CONCLUSION

During the development project the CAN communi-

cation data bus was successfully implemented into the

mobile robot and avionics control system. Its perfor-

mance and reliability satisfy requirements of individ-

ual systems and their standards.

The next step for the future will be implementation

of CAN communication data bus into more complex

robotic systems like a manufacturing robotic system.

Performed tests on SAM proved CAN suitability in

advanced avionics systems.

Type certiﬁcation issued for avionics system with

described communication data bus for use in civil avi-

ation represents one of the ﬁrst realisation of similar

project in the world.

ACKNOWLEDGEMENTS

The analysis of commercial off the shelf data buses

for robotics and airborne applications were acquired

using the subsidization of the Ministry of Industry and

Trade of the Czech Republic, research plan FI–IM/084

”Commercial Off The Shelf (COTS) Databuses De-

velopment for Advanced Avionics and Airborne Sys-

tems (AIR-COTS)”.

Hardware and software development and physical

modules were realized and project was supported by

UNIS, spol. s r.o.

REFERENCES

SZABO

, I., OPLUS

TIL, V. Distributed CAN Based Control

System for Robotic and Airborne Applications. In

ICARV, 5th International Conference on Control, Au-

tomation, Robotics and Vision. Singapore: 2002.

SZABO

, S

., ET AL. Autonomous mobile robot with multi-

processor control system. 7th International Workshop

on Advanced Motion Control, AMC 2002. Maribor,

Slovenia. ISBN 0–7803–7479–7.

SZABO

, S

. Research of local and global navigation methods

for autonomous mobile robot with multiprocessor con-

trol system. IEEE International Conference on Indus-

trial Technology, IEEE ICIT 2003. Maribor, Slovenia.

ISBN 0–7803–7852–0, ISBN 0–7803–7853–9 (CD).

SVAC

INA, D., MIKULA, J. Compliance Report for CRI S-2.

Description of CAN and SAM in Ae270 airplane. Doc-

ument No.: VF-401-AVIO-001, revision: 040322ds.

Unis spol. s r.o.: 3/22/2004.

CAN Speciﬁcation. Bosch, 1995.

<http://www.can.bosch.com>

CANaerospace. Michael Stock, Flight Systems, 2001.

<http://www.stockﬂightsystems.com>

System of Aviation Modules (the SAM) for Ae270 aircraft.

Technical speciﬁcation No.TP-3100-01-01-X, UNIS

Ltd, Brno, Czech Republic. rev. 0, 2001.

CANalyzer and DENalyzer 5.2. The Tools for Comprehen-

sive Network Analysis. Vector, 2005.

<http://www.vector-cantech.com>

DISTRIBUTED CONTROL SYSTEMS BASED ON COTS COMMUNICATION DATA BUS