INTRODUCTION Recently fuzzy logic has found increasing applicability in the field of vehicle control. Applications include automatic transmission, engine control, cruise control, antiskid braking, and air conditioning, among others. This application note focuses on automatic transmission control. AUTOMATIC TRANSMISSION : BASIC MODEL A basic automatic transmission system is shown in Figure 1. Fuzzy logic is employed to infer the best gear selection. The four fuzzy inference unit inputs are sensor based signals from the car itself. Using throttle, vehicle speed, engine speed, engine load, the fuzzy inference unit determines a shift, i.e., gear number, for the car. Figure 1 Automatic Transmission System Definitions of Input/Output Variables To create a fuzzy inference unit, we first need to define labels (membership functions) for input and output variables. Examples of such labels are shown in Figures 2, 3, 4, 5, and 6. The output variable Shift uses singleton membership functions because the TVFI (Truth Value Flow Inference) method is the preferred method of defuzzification. Figure 2 Labels and Membership Functions of Throttle Figure 3 Labels and Membership Functions of Vehicle_Speed Figure 4 Labels and Membership Functions of Engine_Speed Figure 5 Labels and Membership Functions of Engine_Load Figure 6 Labels and Membership Functions of Shift Rules Using labels as defined above, we can write rules for the fuzzy inference unit shown in Figure 1. Rules embody the knowledge base required for decision making. They are represented as English like if-then statements. For example, the following is a rule: IF Throttle is Low and Vehicle_Speed is Low and Engine_Speed is Low and Engine_Load is High THEN Shift is No_1 We can write many such rules to cover the different situations encountered in transmission of power to wheel. The totality of such rules constitutes a fuzzy inference unit for gear selection in an automobile. AUTOMATIC TRANSMISSION : MODIFIED MODEL The performance of the above automatic transmission model is not very good. The gear shifting procedure is implemented without taking into account the driving environment. We, as humans, drive in different "modes" depending on road conditions. For example, we sometimes drive at a constant low gear when negotiating a windy mountainous road. This avoids unnecessary gear shifting, which can add to engine wear and make for a less than smooth ride for passengers. With this in mind, a modified transmission system is shown in Figure 7. We have added an extra input, mode, to the fuzzy inference unit to influence gear shift behavior. This new driving mode can be inferred by fuzzy logic(FIU B) as well. Figure 7 Modified Automatic Transmission System Figure 8 Fuzzy Inference Unit for Driving Mode Figure 8 shows a fuzzy inference unit for inferring driving mode. To create an FIU, we develop rules such as the following: If Vehicle_Speed is Low and Variation_of_Vehicle_Speed is Small and Slope_Resistance is Positive_Large and Accelerator is Medium then Mode is Steep_Uphill_Mode If Vehicle_Speed is Medium and Variation_of_Vehicle_Speed is Small and Slope_Resistance is Negative_Large and Accelerator is Small then Mode is Gentle_Downhill_Mode The driving mode output of FIU B can then be further used to affect the gear shifting procedure. For example, if mode is Steep_Uphill_Mode, a downshift is necessary in order to obtain greater engine power. If mode is Gentle_Downhill_Mode, we also need a lower gear than would be the case for a flat smooth road. The lower gear provides engine braking power. Typical gear selection rules could look as follows: If Mode is Steep_Uphill_Mode then Shift is No_2 If Mode is Gentle_Downhill_Mode then Shift is No_3 COMMENTS In actuality, the inputs to fuzzy inference unit B in Figure 8 could include other factors, such as steering angle, to determine a more accurate driving mode. With steering angle data, we can determine whether or not the vehicle is on a winding road. Gear shifting practices can be quite different on a winding road than on a straight road. Again, fuzzy logic provides us with a powerful tool to deal with complex situations that are intractable using conventional approaches. We simply include additional variables and rules to take into account factors that could improve the behavior of our control system. (Weijing Zhang, Applications Engineer, Aptronix Inc.) For Further Information Please Contact: Aptronix Incorporated 2150 North First Street #300 San Jose, CA 95131 Tel (408) 428-1888 Fax (408) 428-1884 FuzzyNet (408) 428-1883 data 8/N/1 Aptronix Company Overview Headquartered in San Jose, California, Aptronix develops and markets fuzzy logic-based software, systems and development tools for a complete range of commercial applications. The company was founded in 1989 and has been responsible for a number of important innovations in fuzzy technology. Aptronix's product Fide (Fuzzy Inference Development Environment) -- is a complete environment for the development of fuzzy logic-based systems. Fide provides system engineers with the most effective fuzzy tools in the industry and runs in MS-WindowsTM on 386/486 hardware. The price for Fide is $1495 and can be ordered from any authorized Motorola distributor. For a list of authorized distributors or more information, please call Aptronix. The software package comes with complete documentation on how FIDE Application Notes Available: #001 Washing Machine Decision Making, Determining Wash Time #002 Automatic Focusing System Decision Making, Determining Focus #003 Servo Motor Force Control Servo Control, Grasping Object #004 Temperature Control(1) Process Control, Glass Melting Furnace #005 Temperature Control(2) Process Control, Air Conditioner #006 Temperature Control(3) Process Control, Reactor #007 Automatic Transmission Decision Making, Determining Gear Shift FIDE Application Note 007-920929 Aptronix Inc., 1992 Automatic Transmission