AUTOMOTIVE
`ELECTRONICS
`~HANDBOOK
`
`Ronald K. Jurgen Editor in Chief
`
`McGraw-Hill, Inc.
`New York San Francisco Washington, D.C. Auckland Bogota
`Caracas Lisbon London Madrid Mexico City Milan
`Montreal New Delhi San Juan Singapore
`Sydney Tokyo Toronto
`
`BMW1083
`Page 1 of 11
`
`

`

`Library of Congress Cataloging-in-Publication Data
`
`Automotive electronics handbook / Ronald Jurgen, editor in chief.
`p. cm.
`Includes index.
`ISBN 0-07-033189-8
`1. Automobiles--Electronic equipment. I. Jurgen, Ronald K.
`TL272.5.A982 1994
`629.25’49--dc
`
`94-39724
`CIP
`
`Copyright © 1995 by McGraw-Hill, Inc. All rights reserved. Printed in the
`United States of America. Except as permitted under the United States
`Copyright Act of 1976, no part of this publication may be reproduced or dis-
`tributed in any form or by any means, or stored in a data base or retrieval
`system, without the prior written permission of the publisher.
`
`1234567890 AGM/AGM 90987654
`
`ISBN 0-07-033189-8
`
`The sponsoring editor for this book was Stephen S. Chapman, the editing
`supervisor was Virginia Carroll, and the production supervisor was
`Suzanne W. B. Rapcavage. It was set in Times Roman by North Market
`Street Graphics.
`
`Printed and bound by Arcata Graphics/Martinsburg.
`
`McGraw-Hill books are available at special quantity discounts to use as pre-
`miums and sales promotions, or for use in corporate training programs. For
`more information, please write to the Director of Special Sales, McGr’aw-
`Hill, Inc., 11 West 19th Street, New York, NY 10011. Or contact your local
`bookstore.
`
`Information contained in this work has been obtained by McGraw-
`Hill, Inc. from sources believed to be reliable. However, neither
`McGraw-Hill nor its authors guarantee the accuracy or complete-
`ness of any information published herein, and neither McGraw-
`Hill nor its authors shall be responsible for any errors, omissions,
`or damages arising out of use of this information. This work is
`published with the understanding that McGraw-Hill and its authors
`are supplying information, but are not attempting to render engi-
`neering or other professional services. If such services are
`required, the assistance of an appropriate professional should be
`sought.
`
`This book is printed on acid-fi’ee paper:
`
`BMW1083
`Page 2 of 11
`
`

`

`CHAPTER 14
`CRUISE CONTROL
`
`Richard Valentine
`Motorola Inc.
`
`I4.1 CRUISE CONTROL SYSTEM
`
`A vehicle speed control system can range from a simple throttle latching device to a sophisti-
`cated digital controller that constantly maintains a set speed under varying driving conditions
`The next generation of electronic speed control systems will probably still use a separate mod-
`ule (black box), the same as present-day systems, but will share data from the engine, ABS, and
`transmission control systems. Futuristic cruise control systems that include radar sensors to
`measure the rate of closure to other vehicles and adjust the speed to maintain a constant dis-
`tance are possible but need significant cost reductions for widespread private vehicle usage.
`The objective of an automatic vehicle cruise control is to Sustain a steady speed under
`varying road conditions, thus allowing the vehicle operator to relax from constant foot throt-
`tle manipulation. In some cases, the cruise control system may actually improve the vehicle’s
`fuel efficiency value by limiting throttle excursions to small steps. By using the power and
`speed of a microcontroller device and fuzzy logic software design, an excellent cruise control
`system can be designed.
`
`14.1.1 Functional Elements
`
`i~ii¸
`
`The cruise control system is a closed-loop speed control as shown in Fig. 14.1. The key input
`signals are the driver’s speed setpoint and the vehicle’s actual speed. Other important inputs
`are the faster-accel/slower-coast driver adjustments, resume, on/off, brake switch, and engine
`control messages. The key output signals are the throttle control servo actuator values. Addi-
`tional output signals include cruise ON and service indicators, plus messages to the engine
`and/or transmission control system and possibly data for diagnostics.
`
`14.1.2 Performance Expectations
`
`The ideal cruise system features would include the following specifications:
`
`¯ Speed performance: _+0.5 m/h control at less than 5 percent grade, and +1 m/h control or
`vehicle limit over 5 percent grade.
`° Reliability: Circuit designed to withstand overvoltage transients, reverse voltages, and
`power dissipation of components kept to minimum.
`
`14.1
`
`BMW1083
`Page 3 of 11
`
`

`

`CONTROL SYSTEMS
`
`+ 12 V Ignition
`
`I Transient
`
`Protection
`
`Vehicle Speed Sensor~
`
`- On-Off
`- Set-Coast
`- Resume-Aceel
`
`Input Signal
`Conditioning .-~
`(EMI filters) ..~
`(OVP) .~
`
`(Brake/Clutch Switches ;
`
`MCU
`
`Timer
`Input
`Capture
`
`RESET
`
`~MCU Reset I
`
`Control I
`
`Power
`Device
`Drivers
`
`Throttle
`Positioner
`Actuator
`
`J
`
`Data BUS
`Driver
`
`,
`
`Data Bus )
`
`FIGURE 14.1 Cruise control system.
`
`i!i!;
`
`¯ Application options: By changing EEPROM via a simple serial data interface or over the
`MUX network, the cruise software can be upgraded and optimized for specific vehicle
`types. These provisions allow for various sensors, servos, and speed ranges.
`¯ Driver adaptability: The response time of the cruise control can be adjusted to match the
`driver’s preferences within the constraints of the vehicle’s performance.
`¯ Favorable price-to-perfolmance ratio: The use of integrated actuator drivers and a high-
`functionality MCU reduce component counts, increase reliability, and decrease the cruise
`control module’s footprint.
`
`14.1.3 Safety Considerations (Failsafe)
`
`Several safety factors need to be considered for a vehicle speed control design. The most
`basic is a method designed into the throttle control circuit to insure a failsafe mode of oper-
`ation in the event that the microcontroller or actuator drivers should fail. This electronic fail-
`safe circuit shuts off the control servos so that the throttle linkage will be released when the
`brake switch or cruise off switch is activated, no matter the condition of the MCU or servo
`actuator control transistors. (This assumes the actuators are mechanically in good shape and
`will release.)
`Other safety-related items include program code to detect abnormal operating conditions
`and preserving into memory the data points associated with the abnormal condition for later
`diagnostics. Abnormal conditions, for example, could be an intermittent vehicle speed sensor,
`or erratic driver switch signals. A test could also be made during the initial ignition "key on
`time" plus any time the cruise is activated to verify the integrity of the cruise system, with any
`faults resulting in a warning indicator to the driver. Obviously, the most serious fault to avoid
`is runaway acceleration. Continuous monitoring of the MCU and key control elements will
`help minimize the potential for this type of fault.
`
`BMW1083
`Page 4 of 11
`
`

`

`14.2 MICROCONTROLLER REQUIREMENTS FOR CRUISE
`CONTROL
`
`CRUISE CONTROL
`
`14.3
`
`The MCU for cruise control applications requires high functionality. The MCU would include
`the following:
`
`o a precise internal timebase for the speed measurement calculations
`
`o A/D inputs
`PWM outputs’
`
`, timer input capture
`o timer output compares
`o serial data port (MUX port)
`¯ internal watchdog
`, EEPROM
`¯ low-power CMOS technology
`
`14.2.1
`
`Input Signals
`
`The speed sensor is one of the most critical parts in the system, because the inicrocontroller
`calculates the vehicle speed from the speed sensor’s signal to within ~2 m!h.Any speedometer
`cable whip or oscillation can cause errors to be introduced into the speed calculation. An
`averaging routine in the speed calculations can minimize this effect. The speedometer sensor
`drives the microcontroller’s timer input capture line or the external interrupt line. The MCU
`then calculates the vehicle’s speed from the frequency of the sensor signals and the MCU
`internal timebase. The vehicle’s speed value is continually updated and stored into RAM for
`use by the basic speed control program. Speed sensors traditionally have been a simple ac
`generator located in the transmission or speedometer cable. The ac generator produces an ae
`voltage waveform with its frequency proportional to the sensor’s rpm and vehicle speed. Opti-
`cal sensors in the speedometer head can also be incorporated. Usually the speed sensor pro-
`duces a number of pulses or cycles per km or mile. With the increasing ABS system usage, a
`backup speed sensor value could be obtained from the ABS wheel speed sensors. The ABS
`speed data could be obtained by way of a MUX network.
`The user command switch signals could either be single MCU input lines to each switch
`contact or a more complex analog resistor divider type to an A/D input line. Other input sig-
`nals of interest to the cruise system program would be throttle position, transmission or clutch
`status, A/C status, actuator diagnostics, engine status, etc., which could be obtained over the
`MUX data network.
`
`14.2.2 Program Flow
`
`The microcontroller is programmed to measure the rate of vehicle speed and note how much,
`and in which direction, the vehicle speed is drifting. The standard PI (proportional-integral)
`method produces one output signal p that is proportional to the difference between the set-
`speed and actual vehicle speed (the error value) by a proportional gain block Kp. Another sig-
`nal i is generated that ramps up or down at a rate set by the error signal magnitude. The gains
`of both Ki and Kp are chosen to provide a quick response, but with little instability. In effect,
`the PI system adds up the error rate over time, and, therefore, if an underspeed condition
`occurs as in a long uphill grade, the error signal will begin to greatly increase to try to com-
`pensate. Under level driving conditions, the integral control block Ki will tend toward zero
`
`L
`
`BMW1083
`Page 5 of 11
`
`

`

`14.4
`
`CONTROL SYSTEMS
`
`Set Speed Value -
`
` rrorVaue ¯
`
`~1 Proportiona~
`
`[ Gain, Ki
`
`iKp
`
`’ [
`
`Dri trai
`
`Actual Vehicle Speed Value
`
`ISpeed Senso~
`
`FIGURE 14.2 PI speed error control.
`
`because there is less error over time. The vehicle’s weight, engine performance, and rolling
`resistance all factor in to determine the PI gain constants. In summary, the PI method allows
`fast response to abrupt grades or mountains and stable operation under light grades or hills.
`Figure I4.2 shows the traditional PI cruise control diagram,
`
`14.2.3 Output Controls
`
`When the error signal has been computed, an output signal to the serve actuators is generated
`to increase, hold, or decrease the throttle position. The serve is updated at a rate that is within
`the servo’s mechanical operating specifications, which could be several milliseconds.The error
`signal can be computed at a much faster rate and, therefore, gives extra time for some aver-
`aging of the vehicle speed sensor signal.
`Throttle positioning is traditionally either a vacuum type serve or motor.The vacuum sup-
`ply to the vacuum serve/actuator is discharged as a failsafe measure whenever the brake sys-
`tem is engaged in addition to the normal turn-off of the actuator driver coils. Electric serve
`type motors require more complex drive electronics and some type of mechanical failsafe
`linked backed to the brake system.
`
`14.3 CRUISE CONTROL SOFTWARE
`
`The cruise error calculation algorithm can be designed around traditional math models such
`as PI or fuzzy logic.
`
`14.3.1 Fuzzy Logic Examples
`
`Fuzzy logic allows somewhat easier implementation of the speed error calculation because its
`design syntax uses simple linguistics. For example: IF speed difference negative and
`THEN increase throttle slightly.
`The output is then adjusted to slightly increase the throttle. The throttle, position update
`rate is determined by another fuzzy program which looks for the driver’s cruise performance ......
`request (slow, medium, or fast reaction), the application type (small, medium, or large
`size), and other cruise system factory preset parameters. Figure 14.3 shows one part of a fuzzy
`logic design for computing normal throttle position. Other parts wou!d compute the effects of
`other inputs, such as resume, driver habits, engine type, and the like.
`Other program design requirements include verification that the input signals fall within
`expected boundaries. For example, a broken or intermittent speed sensor could be
`
`BMW1083
`Page 6 of 11
`
`

`

`CRUISE CONTROL
`
`14.5
`
`Vehicle Speed to Setpoint Error
`
`NMed NSml NVSMLPVSML PSml PMed
`
`\
`
`Accumulative Vehicle Speed Error Rate
`
`Large
`Deaccel
`
`Small VSML VSML Smd Medium
`Medium
`Daaccel DeaccelDeaccelAccel Accel
`Acc~l
`
`Large
`Accel
`
`/
`/
`
`64 RULES
`IF V~hic~Spe~d Error NLJuge AND Deac~a! rate L~ge THEN Throt~epos L.mge Increase
`IF VehideSpeed Error NLarge AND Deaccel rate Medium THEN "~rottlepos Medium Increase
`
`IF Vehlc/eSpeed Error NVsrnall AND Deaccel rate Vsma]l THEN Throttlepos SmeJI Increase
`
`IF VehideSpeed Error PLarge AND Acoel rate Large THEN Throttlepos Large De~ease
`
`Throttle Position Step Size
`
`Vehicle Speed
`(, Input Signal Conditioning )
`I
`
`I averaging, noise filters, etc. I
`
`scaling, out of range test
`
`Output Signal
`Conditionin~l )
`I
`
`I scaling, interface control, etc. ]
`
`I
`~ Throttle Control Output )
`
`Medium
`Decmane
`
`Large
`Decrea_~e
`
`+I! 0,. r "
`
`FIGURE 14.3 Fuzzy speed error program llow.
`
`Step %
`
`Small NO SmaJl Me,urn Large
`[:)ecmase Change Increase Incmasa
`tanreasa
`
`I
`
`A heavily loaded vehicle with a small engine may not be able to maintain a high setpoint
`speed up a steep grade, and the cruise control needs to be disengaged to protect the engine
`from sustained full-throttle operation under a heavy load. This could be preset to occur 20
`percent below tile setpoint speed. Another program can test the vehicle speed to resume set-
`point speed and prevent unsafe acceleration under certain conditions. For example, if a high-
`performance vehicle (>200-kW or 268-hp engine) has a setpoint speed of 125 km/h (78 mi/h),
`and drives from the freeway into heavy city traffic doing 48 km/h (30 mi/h) and the vehicle’s
`
`BMW1083
`Page 7 of 11
`
`

`

`CONTROL SYSTEMS
`
`driver fortuitously hits the cruise resume switch at this low speed, the cruise control invokes a
`near full-throttle action, and an accident is likely. A fuzzy design can limit the acceleration
`upon resume using simple rules such as IF resume and big speed error, THEN increase throt-
`tle slightly.
`
`i!iiiii i i! ii!
`
`14.3.2 Adaptive Programming
`
`The response time and gain of the cruise system can be adjusted to match individual drivers.
`For example, some drivers may prefer to allow the vehicle to slow down somewhat when
`climbing a grade and then respond quickly to maintain a setspeed; other drivers may prefer a
`constant speed at all times, while still other drivers may prefer a very slow responding cruise
`system to maximize fuel efficiency. The cruise system can be adapted either by a user selection
`switch (slow, medium, fast) or by analyzing the driver’s acceleration/deacceleration habits
`during noncruise operation. Once these habits are analyzed, they can be grouped into the
`three previously mentioned categories. One drawback of a totally automatic adaptive cruise
`system is when various drivers with vastly different driving preferences operate the vehicle on
`file same trip. The cruise system would have to be "retrained" for each driver.
`
`i
`
`14.4 CRUISE CONTROL DESIGN
`
`Many of the required elements of a cruise control can be integrated into one single-chip MCU
`device. For example, the actuator drivers can be designed in the MCU if their power require-
`ments are on the low side.
`
`14.4.1
`
`Automatic Cruise System
`
`Figure 14.4 shows an experimental system design for a cruise control based upon a semieus-
`tom 8 or 16-bit single-chip MCU that incorporates special high-power output driver elements
`and a built-in voltage regulator.
`
`14.4.2 Safety Backup Examples
`
`The design of a cruise control system should include many safeguards:
`
`¯ A test to determine vehicle speed conditions or command inputs that do not fall within the
`normal conditions for operation of the cruise control function.
`¯ A test to determine if the vehicle speed has decreased below what the cruise routine can
`compensate for.
`¯ Speed setpoint minimums and maximums (30 km/h min to 125 km/h max, for example) are
`checked and, if exceeded, will cause the cruise function to turn off.
`¯ Speedometer cable failure is detected by checking for speed sensor electrical output pulses
`over a 100-ms time period and, if these pulses are absent, the system is disengaged.
`¯ Software program traps should also be scattered throughout the program and, if memory
`permits, at the end of each program loop. These will catch an out-of-control program and
`initiate a vector restart.
`
`BMW1083
`Page 8 of 11
`
`

`

`+ 12 V Ignition
`
`Transient
`Protection
`
`CRUISE CONTROL
`
`14,7
`
`Vehicle Speed Sensor~==~-
`
`I S elector Switches
`
`- On-Off
`- Set-Coast
`- Resume-Accel
`
`Throttle Position Sensor}-~-
`
`Semi-Custom
`MCU
`
`( Brake/Clutch Switche@F
`
`FIGURE 14.4 Automatic cruise control.
`
`"--~Data Bus )
`
`Positioner
`Actuator
`
`~ Throttle
`
`14.4.3 EMI and RFI Noise Problems
`
`As with any electronic design, consideration must be given to suppressing RFI (radio fre-
`quency interference) from the circuit, besides minimizing effects of external EMI (electro-
`magnetic interference) and RFI to the circuit’s normal operation. It is not uncommon that the
`circuit must operate in RF fields up to 200 V/m intensity. This requires careful layout of the
`module’s PCB (printed circuit board) and RF filters on all lines going in or out of the module.
`The module case may even have to contain some type of RF shielding. Minimizing generated
`RFI from the cruise circuit can be accomplished by operating the MCU’s crystal oscillator at
`a minimal power level (this is controlled mostly by the MCU internal design), careful PCB
`trace layout of the MCU oscillator area, metal shielding over the MCU, ground planes on the
`PCB under the MCU, and setting the actuator switching edge transition times to over 10 ms.
`(See Chaps. 27 and 28.)
`
`14.5 FUTURE CRUISE CONCEPTS
`
`Several research projects are underway to develop a crash avoidance system that could be
`interconnected with a cruise system. The development of a low-cost distance sensor that can
`measure up to a few hundred meters away with a tight focal point in all weather conditions
`is proving to be a challenge.When a practical vehicular distance sensor is available, the cruise
`control can be programmed to maintain either constant speed or constant distance to
`another vehicle. Other methods of cruise control could include receiving a roadside signal
`that gives an optimum speed value for the vehicle when travelling within certain traffic con-
`trol areas.
`
`BMW1083
`Page 9 of 11
`
`

`

`CONTROL SYSTEMS
`
`Road Conditions Integration with IVHS
`
`V
`
`The IVHS (Intelligent Vehicle-Highway System) network ~ay be a more practical approach
`to setting optimum cruise speed values for groups of vehicles. The IVHS can monitor road
`conditions, local weather, etc., and broadcast optimal speed data values for vehicles in its zone
`(See Chap. 29.)
`
`GLOSSARY
`
`Analog input Sensors usually generate electrical signals that are directly proportional to the
`mechanism being sensed.The signal is, therefore, an~!gg or can vary from a minimum limit to
`a maximum limit. Normally, an 8-Nt MCU A/D input using a 5-V reference, the analog input
`resolution is I bit, which is 1/256 of 5 V o,r 010193 V.
`
`Defnzzitication The process of translating output grades to analog output values.
`
`Fuzzification The process of translating analog inP}lt values to input memberships or labels
`
`Fuzzy logic Software design based upon a reasoning model rather than fixed mathematical
`algorithms.A fuzzy logic design allows the sYs!em engineer to participate in the software design
`because the fuzzy language is ling u!stic and ,built upon easy-to-comprehend fundamentals.
`
`Inference engine The interna! software program that produces output values through fuzzy
`rules for given input values. The inference’process invol,¢es three steps: fuzzification, rule eval-
`uation, and defuzzification.
`
`Input memberships The input signal or sensor range is divided into degrees of membership,
`i.e., low, medium, high or cold; cool, c0mfortab!e, warm, hot. Each of these membership labels
`is assigned numerical values Or grades.
`
`Output memberships The output signa! is divided into grades such as off, slow, medium, fast,
`and full-on. Numerical values are assigned t0each grade. Grades can be either singleton (one
`value) or Mandani (a range of values per grade).
`
`Rule evaluation Output values are comPUted pel the input memberships and their relation-
`ship to the output memberships. The number Of rules is usually set by the total number of
`input memberships and the tota! number~ of OUtput memberships. The rules consist of IF
`inputvarA is x, AND inputvarB is y, THEN outvar is z.
`
`Semicustom MCU An MCU (microcontrol!er unit) that incorporates normal MCU ele-
`ments plus user-specified periphera! devices such as higher-power port outputs, special timer
`units, etc. Mixed semiconductor te¢hnglogies, S qch as high-density CMOS (HCMOS) and
`bipolar analog, are available in a semicust0m MCU. Generally, HCMOS is limited to 10 V,
`whereas bipolar-analog is usat!le to 60 V.
`
`BIBLIOGRAPHY
`
`Bannatyne, R.,"Fuzzy logie--A new approach to embedded
`Design Concept, DC410,1992.
`
`Motorola Semiconductor Application Note, AN1050,: 1989.
`
`BMW1083
`Page 10 of 11
`
`

`

`CRUISE CONTROL
`
`14.9
`
`Chaudhuri, et al., "Speed control integrated into the powertrain computer," New Trends in Electronic
`Management and Driveline Controls, SAE SP-653, 1986, pp. 65-72.
`Hosaka, T., et al., "Vehicle control system and method therefore," U.S. Patent 4809175, May 29,1990.
`Hosaka, T., et al., "Vehicle control system," U.S. Patent 4930084, Feb. 28,1989.
`Mamdani, E. H., "Application of fuzzy logic to approximate reasoning using linguistic synthesis," 1EEE
`Transactions on Computers, C-26-12,1977, pp. 1182-1191.
`Ribbens, W., ~’Vehicle Motion Control," UnderstandingAutomotive Electronics, 4th ed., 1992, pp. 247-257.
`Takahashi, Hioshi, "Automatic speed control device using self-tuning fuzzy logic," IEEE Workshop on
`Automotive Applications of Electronics, 88THO321,1988, pp. 65-71.
`Sel~ Kevin, "Designing with fuzzy logic," 1EEE Spectrum, Nov. 1990, pp. 42-44, 105.
`Sibigtroth, J., "Implementing fuzzy expert rules in hardware," AI Expert, April 1992.
`Stefanides, E. J., "Cruise control components packaged as one unit," Design News, Oct. 1, 1990, pp.
`162-163.
`Zadeh, L. A., "Fuzzy sets, information and control," vol. 8,1965, pp. 338-353.
`
`ABOUT THE AUTHOR
`
`Richard J. Valentine is a principal staff engineer at Motorola SPS in Phoenix, Ariz. His present
`assignments include engineering evaluation of advanced semiconductor products for emerging
`automotive systems. He holds two patents and has published 29 technical articles during his 24
`years at Motorola.
`
`BMW1083
`Page 11 of 11
`
`