Major Project End-Term Report On Automatic Landing Controller using Artificial Intelligence Technique

Submitted by: A.K. Dharma Teja (R640209002) Harpreet Singh Sidhu (R640209019) N. Adhiyaman (R640209025) B.Tech. Avionics Engineering (2009-13)

Under the Supervision of: Prof. M. Raja

Department of Aerospace Engineering College of Engineering Studies (COES) University of Petroleum & Energy Studies (UPES), Dehradun October, 2012

FUZZY CONTROL SYSTEM FOR UAV LANDING TASK

ABSTRACT
The unmanned aerial vehicle (UAV) has made its wayquickly and decisively to the forefront of aviation technology.Opportunities exist in a broadening number of fields for theapplication of UAV systems as the components of thesesystems become increasingly lighter and more powerful.UAVs provide a cheap and safe alternative against to manned systems and often provide a far greater magnitude of capability. Previously studied controller of the UAVis the use of dynamic inversion with neural networks.Dynamic inversion is a control law design methodologythat works to cancel out the vehicle dynamics. However,since the accurate dynamics are required, the neuralnetwork method must be trained online to adapt to themodeling errors. But in this project, well try to approach byanother method, fuzzy control system.

OBJECTIVE
The project mainly focuses on designing the landing controller of UAV by Fuzzy Control System to perform simple attitude control.

FUZZY CONTROL SYSTEM

Fuzzy logic is widely used in machine control. The term refer to the logic involved that can deal with fuzzy conceptsconcepts that cannot be expressed as "true" or "false" but rather as "partially true". Although genetic algorithms and neural networks can perform just as well as fuzzy logic in many cases, fuzzy logic has the advantage that the solution to the problem can be cast in terms that human operators can understand, so that their experience can be used in the design of the controller. This makes it easier to mechanize tasks that are already successfully performed by humans.

WHY FUZZY LOGIC?

Normally in logic we have a series of statements which are either true or false. In this context, the statement the temperature is 25 degrees Celsius is an objective one and is either true or false.However for many situations the answer is more like Errr .For example, on a pleasant summers day the statement the temperature is too high is neither true nor false. The statement is qualitive one it represents an opinion rather than an objective fact. For example it needs to be a bright sunny day on the beach for me to feel warm. On the other hand,I could mention some visiting scientists at control system principles who feel comfortable in a snow storm on top of a mountain. Do you see what I mean? There is no certainty to the situation- it depends upon the context. Fuzzy logic deals with uncertainty in engineering by attaching degree of certainty to the answer to a logical question. Why should this be useful? The answer is commercial and practical. Commercially, fuzzy logic has been used with great success to control machines and consumer goods. In the right applications fuzzy logic systems are simple to design, and can be understood and implemented by non specialists in control theory. In most cases someone with an intermediate technical background can design a fuzzy logic controller. The control system will not be optimal but it can be acceptable/. Control engineer also use it in applications where the on board computing is very limited and adequate control is enough. Fuzzy logic is not the answer for all technical problems but for control problems where simplicity ands speed of implementation is important then fuzzy logic is a strong candidate. A cross section of applications that have successfully used fuzzy control includes: Environmental Control Air conditioners Humidifiers

Domestic Goods Washing Machines Vacuum cleaners Toasters Microwave ovens Refrigerators

Consumer electronics Television Photocopiers Still and video cameras

Hi-fi systems

Automotive systems Vehicle climate control Automatic gear box Four-wheel steering Seat/mirror control systems

This is an impressive list, and gives an idea of the key application areas in general you will not find a fuzzy controller in a safety critical application, unless the practical and theoretical performance has been completely studied

METHODOLOGY
Theoretical study of Fuzzy Logic Toolbox. Study of MATLAB SIMULINK for Simulation.

PROGRESS TILL DATE The Fuzzy Inference Diagram

The fuzzy inference diagram is the composite of all the smaller diagrams presented so far in this section. It simultaneously displays all parts of the fuzzy inference process you have examined. Information flows through the fuzzy inference diagram as shown in the following figure.

Figure 1: Fuzzy Inference System

In this figure, the flow proceeds up from the inputs in the lower left, then across each row, or rule, and then down the rule outputs to finish in the lower right. This compact flow shows everything at once, from linguistic variable fuzzification all the way through defuzzification of the aggregate output. The design of both Mamdani and Sugeno-type fuzzy controller has been done according to [a], this means that fuzzy rules, fuzzy membership functions, have been modeled accordingly. The system proposed in [a] is Mamdani-type fuzzy inference system. Although it is not clearly stated, we can conclude this based on output control force membership functions which are all distributed. The simplifications regarding time and mass are also consistent with [a]: t = 1.0 (s) m = 1.0 (lb) Since we have these simplifications then vertical velocity is calculated as in [a]: + (a)

Vertical velocity changes only if control force is applied. Negative control force directs the aircraft to the ground while positive control force directs the aircraft from the ground. Height changes according to [a]: = + (b)

If current vertical velocity is negative, than aircraft will descend, if it is positive it will increase altitude, and if it is zero, then aircraft will remain at the same altitude. All three different types of controllers have two inputs: height in feet and vertical velocity in feet per second. The output is also the same variable for all three types of controllers: control force in pound-force. Control force can change current vertical velocity, and change of vertical velocity, change rate of descent or climb, thus affects current height.

Figure 2: Control System Block

Figure 3: Fuzzy System Using Matlab

Membership Functions
A membership function (MF) is a curve that defines how each point in the input space is mapped to a membership value (or degree of membership) between 0 and 1. The input space is sometimes referred to as the universe of discourse, a fancy name for a simple concept.

Fuzzy sets describe vague concepts (e.g., fast runner, hot weather, weekend days). A fuzzy set admits the possibility of partial membership in it. (e.g., Friday is sort of a weekend day, the weather is rather hot). The degree an object belongs to a fuzzy set is denoted by a membership value between 0 and 1. (e.g., Friday is a weekend day to the degree 0.8). A membership function associated with a given fuzzy set maps an input value to its appropriate membership value.

Membership functions for height are all trimf (triangle membership function) type, and are symmetrical in shape like the membership functions for vertical velocity and Mamdani output membership functions.

Figure 4: Landing controller in fuzzy toolbox

Control force and vertical velocity use trimf and tramf (trapezoid membership functions).Membership functions for Mamdani and Sugeno-type inference systems are the same for input variables. Sugeno-type controller uses different membership functions for the output control force. All Sugeno output membership functions are linear type. The parameters for these membership functions are manually picked.

Figure 5: Membership function for velocity

Figure 6: Membership function for control force

Fuzzy Rules
Fuzzy sets and fuzzy operators are the subjects and verbs of fuzzy logic. These if-then rule statements are used to formulate the conditional statements that comprise fuzzy logic. A single fuzzy if-then rule assumes the form if x is A then y is B Where A and B are linguistic values defined by fuzzy sets on the ranges (universes of discourse) X and Y, respectively. The if-part of the rule "x is A" is called the antecedent or premise, while the then-part of the rule "y is B" is called the consequent or conclusion. The Rule Viewer displays a roadmap of the whole fuzzy inference process. It is based on the fuzzy inference diagram described in the previous section. You see a single figure window with 10 plots nested in it. The three plots across the top of the figure represent the antecedent and consequent of the first rule. Each rule is a row of plots, and each column is a variable. The rule numbers are displayed on the left of each row. You can click on a rule number to view the rule in the status line. Fuzzy rules are usually represented using the compact graphical form Fuzzy Associative Memory table (FAM table). FAM table for this problem has two dimensions since there are two inputs (height and vertical velocity). Our FAM table is the same as in [1] since we used the same set of rules during evaluation of controllers (Table 1.). Height L M S NZ DL Z US UL UL DS DS Z US UL Velocity ZERO DL DS Z Z Table 1 US DL DL DS DS UL DL DL DL DS

In the FAM table each column represents membership function for vertical velocity. DL is Down Large, DS is Down Small, US is Up Small and UL is Up Large. Each row is marked by membership functions of the height input. L is Large, M is Middle, S is Small and NZ is Near Zero. Values in the table are corresponding control force outputs for each possible (in this model) combination of vertical velocity and height. Z is Zero,

DS is Down Small, DL is Down Large and UL is Up Large. FAM table shows that there are 20 rules, which is correct since there are four membership functions for height and five membership functions for vertical velocity. Same rules are used both by Mamdani and Sugenotype controller. ANFIS controller does not use this rule base; it constructs its own rule base

Figure 7: Rule set

Control Surface
Control surface graphically represents all possible inputs and outputs, which is in this case threedimensional, since we have two inputs and one output. Control surface for this problem shows graph of control force in function of vertical velocity and height, for all values in range of both inputs (Figure 7.). We can see possible landing situations from this surface. For example, the graph shows that the control force is large for negative large vertical velocity and small height.

Figure 8: Surface View

FUTURE WORK
Future work includes to implement the Fuzzy controller for the possible movements of the UAV (Forward, Sideand heading), in order to perform a total autonomous landing task in all axis. Performing the simulation in flight gear.

CONCLUSION
An approach of the application of landing task using Fuzzy Logic for UAV is presented in this project.

REFERENCES
1. Olivares, M., Campoy, P., Correa, J., Martinez, C., Mondragon, I.: Fuzzy control system navigation using priority areas. In: Proceedings of the 8th International FLINS Conference, pp. 987996. Madrid, Spain (2008) 2. Olivares, M., Madrigal, J.: Fuzzy logic user adaptive navigation control system for mobile robots in unknown environments. Intelligent Signal Processing, 2007. WISP 2007. IEEE International Symposium on pp. 16 (2007). DOI 10.1109/WISP.2007.4447633 3. Olivares-Mendez, M.A., Campoy, P., Martinez, C., Mondragon, I.: Visual servoing using fuzzy controllers on an unmanned aerial vehicle. Eurofuse workshop 09, Preference modelling and decision analysis (2009)