This project deals with the analysis of vehicle speed and overtaking distance of another vehicle to prevent the accident. The most common reason for accident is misjudging the correlation of speed of our vehicle and length of the overtaking vehicle. Here we are calculating the length of up going vehicle using ultrasonic sensor. The sensor output is based on two parameters namely length of the overtaking vehicle and present speed of our vehicle. Based on the sensor output, condition will be displayed in the LCD screen which was already programmed using Keil software.
The program is embedded in a chip in the form of Embedded C using Keil software for various sensor outputs. Based on these two parameters the driver has to decide. For this analysis we are using algorithm in neural network to make decision making process for various conditions and to display the best result in the LCD screen to the driver. Image analysis technique is used to ensure whether the object is a vehicle or any other living thing (cow, dog).
Thus by calculating the length of the overtaking vehicle and correlating with the speed of our vehicle accidents are avoided using this methodology.
The inter-vehicle spacing or headway affects both safety and highway capacity. For collision free vehicle following, the spacing should be large enough in order to guarantee no collisions during all possible vehicle maneuvers. Lane changing/merging accidents consist of various types of vehicle collisions, such as rear-end collisions, single vehicle road departure accidents, side-wipe, and angle collisions. Lane changing/merging collisions constituted about 4. % of all police-reported collisions in 2010, and accounted for about 0. 5% of all fatalities. Although the lane change crash problem is small relative to other types of crashes and does not account for a high percentage of traffic fatalities, this crash type is responsible for one-tenth of all crash-caused traffic delays often resulting into congestion. Traffic delays and congestion in general increases travel time and has a negative economic impact. Due to the advances in technology, automobiles are available at cheaper prices leading to increase in the rate of traffic at a much greater pace as compared to the addition of road capacity of the highways.
The Essay on Accidents: Traffic and Underground Roads
Road accidents claim far too many lives. What can governments and individuals do to reduce this problem? Throughout this century, road accidents happen quite frequently around the world. As a result too many damages are caused and lives are claimed by these accidents. In my opinion, there are many things that governments and individuals should do in order to reduce this problem. Some of them I ...
Essential maneuvers like lane change, merging and braking etc in high density scenario cause traffic non equilibrium may also cause accidents, congestion etc. To obviate such conditions, forewarning of such conditions is required. Forewarning systems must be able to predict the onset for such traffic situations or suggest remedial action once the situation has precipitated due to an essential maneuver. Lane change is an essential action by a vehicle which perturbs the traffic and may cause collisions between vehicles. Prediction of the feasibility of a safe lane change and the possibility of collision during lane change is desired for smooth flow of traffic and collision avoidance.
Neural network based modeling has attracted increasing attention they are able to learn the traffic conditions and driver behavior at the time of lane change more precisely and realistically than other existing methods. However, the accuracy and validity depends mainly on the drivers’ behavior during lane change. Majority of the accidents on highways take place during lane change maneuvers which obstruct the traffic flood and create an overload in addition to loss of precious human life. The advances in technology have made it possible to simulate the complex models of highway traffic and give the warning for safe driving on the highways during the lane changing and overtaking. CHAPTER 2 LITERATURE REVIEW Naranjo et al. (1) have proposed an evolution of the simple automatic lane-change system that upgrades autonomous vehicles to use this maneuver to perform complete overtaking operations. This paper also proposed a two-level architecture. The low level consists of two fuzzy steering controllers- one for path tracking other for lane changes. Hossein jula et al. ,(2) have proposed in this paper, the problem of safe lane changing and merging maneuvers in highway systems. By analyzing the kinematics of the vehicles involved in a lane changing or merging scenario, we present a general algorithm to calculate whether a particular lane changing/merging maneuver is safe, i.
The Term Paper on Film Techniques against the Theme of “Change”
How are we made aware of the filmmaker’s attitude towards change? Refer to three specific episodes from the film. (excl. concl. stages)In Pleasantville, the filmmaker, Gary Ross, conveys his attitude towards change through the characters of David and Jennifer who are transported into the 1950s sitcom “Pleasantville”. He doesn’t necessarily demonstrate change to bear a ...
e. free of collisions. Moreover, we present a general algorithm for calculating the Minimum longitudinal Safety Spacings (MSS), that is, given a particular lane change/merging scenario, we calculate the minimum longitudinal spacings that the vehicles should initially have so that no collision takes place during the lane changing/merging maneuver. Datta N. Godbole et al. ,(3) have proposed in this paper presents distributed hybrid control designs for automated vehicle lane changes. Lane changes can be safe without coordination but in the general they will take a long time. Likewise they can be efficient without coordination but in general will be un safe.
We show that distributed controls, that us communication to coordinate the motion of the principal vehicle undergoing lane changes with its neighbours, enable maneuevers that are both safe and efficient. Javier Diaz Alonso et al. ,(4) his proposed an aid system based on image processing to help the driver in these situations. The main purpose of an overtaking monitoring system is to segment the rear view and track the overtaking vehicle. We address this task with an optic-flow-driven scheme, focusing on the visual field in the side mirror by placing a camera on top of it. When driving a car, the ego-motion optic-flow pattern is very regular, i. e. all the static objects (such as trees, buildings on the roadside, or landmarks) move backwards. An overtaking vehicle, on the other hand, generates an optic-flow pattern in the opposite direction, moving forward toward the vehicle. This well-structured motion scenario facilitates the segmentation of regular motion patterns that correspond to the overtaking vehicle. Our approach is based on two main processing stages: First, the computation of optical flow in real time uses a customized digital signal processor (DSP) particularly designed for this task and, second, the tracking stage itself, based on motion pattern analysis, which we address using a standard processor.
The Essay on Government Regulation of Climate Change
For all nations, adapting to climate change has become a key issue. Governments are mandated to formulate policies aimed at safe guarding lives and livelihoods for their people and at the same time attain their development goals. As cities continue to grow, more people moving to urban centers, and nations working to become industrialized, climate change is here to stay, all that can be done is ...
e present a validation benchmark scheme to evaluate the viability and robustness of the system using a set of overtaking vehicle sequences to determine a reliable vehicle-detection distance. Plamen Petrov et al. ,(5) his proposed in this paper that, presents a nonlinear adaptive controller for a two-vehicle automated overtaking maneuver. Various control laws for different maneuvers of automated vehicles has been designed and analyzed during the last two decades. The ultimate goal in automating the driving process is to reduce accidents caused by human error and increase safety. While considerable research work has been reported on lane keeping and lane change maneuver, the problem of automated overtaking has attracted less attention.
In contrast to lane keeping and lane change maneuvers, the overtaking is a composition of three consecutive maneuvers: lanechange, followed by trajectory tracking and again a lane change, which have to be planned and coordinated. A two layer controller architecture for overtaking maneuver. The lower level consists of two fuzzy steering controllers for path tracking and lane change, and the high-level is to evaluate the necessity and possibility of overtaking, and to switch between the low-level controllers. Fenghui Wang et al. ,(6) have proposed an overtaking control method based on the estimation of the conflict probability. This method uses the conflict probability as the safety indicator and completes overtaking by tracking a safe conflict probability.
The conflict probability is estimated by the future relative position of intelligent vehicles, and the future relative position is estimated by using the dynamics models of the intelligent vehicles. The proposed method uses model predictive control to track a desired safe conflict probability and synthesizes decision making and control of the overtaking maneuver. The effectiveness of this method has been validated in different experimental configurations, and the effects of some parameters in this control method have also been investigated. Umar Farooq et al. ,(7) have proposed in this paper that, design of a low cost autonomous vehicle based on neural network for navigation in unknown environments is presented.
The Essay on Sound Waves Wave Intensity Ear
Sound Waves Waves are disturbances that travel through a space while changing its matter. A sound wave is what allows us to hear sounds. It is created by vibrations, which are made by the movement of matter. Sound waves must travel through a solid, a liquid, or a gas. The more tight the particles, the faster the wave will travel. This would mean that a sound wave travels fastest through solids and ...
The vehicle is equipped with four ultrasonic sensors for hurdle distance measurement, a wheel encoder for measuring distance travelled, a compass for heading information, a GPS receiver for goal position information, a GSM modem for changing destination place on run time and a non volatile RAM for storing waypoint data; all interfaced to a low cost AT89C52 microcontroller. The microcontroller processes the information acquired from the sensors and generates robot motion commands accordingly through neural network. The neural network running inside the microcontroller is a multilayer feed-forward network with back-propagation training algorithm. The network is trained offline with tangent-sigmoid as activation function for neurons and is implemented in real time with piecewise linear approximation of tangent-sigmoid function. Results have shown that upto twenty neurons can be implemented in hidden layer with this technique.
The vehicle is tested with varying destination places in outdoor environments containing stationary as well as moving obstacles and is found to reach the set targets successfully. Umar Farooq at el. ,(8) have proposed in this paper that, a low cost implementation of a multilayer feed-forward network with back propagation training algorithm for a car like robot is realized using a single readily available 89C52 microcontroller. The network is trained offline by using MATLAB neural network toolbox with tangent-sigmoid being the firing function of neurons. Because linear functions require much lesser computation power than nonlinear functions, tangent-sigmoid function is approximated as piecewise linear function for implementation in microcontroller. This helps in reducing computational time.
Also the converged floating point weights are converted into integers and saved as integers thereby reducing the storage space. Thus both computational time and storage space are reduced to improve the efficiency and speed of the robotic system. CHAPTER 3 COMPONENT DESCRIPTION 3. 1 ULTRASONIC SENSOR Ultrasonic sensors (also known as transceivers when they both send and receive) work on a principle similar to radar or sonar which evaluate attributes of a target by interpreting the echoes from radio or sound waves respectively. Ultrasonic sensors generate high frequency sound waves and evaluate the echo which is received back by the sensor. Sensors calculate the time interval between sending the signal and receiving the echo to determine the distance to an object.
The Essay on Sound Waves Travel In Different Material
A sound wave is a disturbance. When it travels through air, it bounces the air molecules around and they vibrate. They then hit other molecules and cause a chain reaction. In a different material, such as metal, sound actually travels faster. this is because the molecules are much more tightly packed (water is not dense because the molecules just roll over each other, and air is even less dense, ...
This technology can be used for measuring: wind speed and direction (anemometer), fullness of a tank and speed through air or water. For measuring speed or direction a device uses multiple detectors and calculates the speed from the relative distances to particulates in the air or water. To measure the amount of liquid in a tank, the sensor measures the distance to the surface of the fluid. Further applications include: humidifiers, sonar, burglar alarms and non-destructive testing. Systems typically use a transducer which generates sound waves in the ultrasonic range, above 18,000 hertz, by turning electrical energy into sound, then upon receiving the echo turn the sound waves into electrical energy which can be measured and displayed.
The sound waves are generated as recurrent changes in electrical voltage occurring at an ultrasonic rate, since it is above the audible range. Ultrasonic vibrations of lower frequencies act in essentially the same manner as audible sound waves. Ultrasonic waves of the higher frequencies behave somewhat like light waves. Ultrasonic vibrations have two basic characteristics: a. They are reflected by discontinuities occurring in the medium thorough Which they are travelling and, b. They tend to travel in a straight line, as do light waves, due to the Shortness of the wave length employed. 3. 1. 1. SOUND AND ULTRASOUND PRINCIPLES Sound is a mechanical vibration transmitted by an elastic medium.
The range of frequencies that humans can hear is approximately between 20 Hz and 20,000 Hz. This range is by definition the audible spectrum and varies by individual and generally reduces with age. The ear is most sensitive to frequencies around 3,500 Hz. Sound above 20,000 Hz is known as ultrasound, and sound below 20 Hz as infrasound. 3. 1. 2. DIAGRAM OF ULTRASONIC SENSOR [pic] Figure 3. 1. ULTRASONIC SENSOR 3. 1. 3. WORKING OF ULTRASONIC SENSOR [pic] Figure 3. 2 WORKING OF ULTRASONIC SENSOR Where, MCU- MICROCONTROLLER UNIT Tx – TRANSMITTER Rx – RECEIVER 3. 1. 4. TRANSDUCER An ultrasonic transducer is a device that converts energy into ultrasound, or sound waves above the normal range of human hearing.
The Essay on The Sights and Sounds of High School
Walking down the hall today, I heard many interesting sounds. A) The sound of young freshmen screaming their high, annoying screeches echoes throughout the halls. B) Students take part in heated debates, as if they understood any politics, about the presidential election everywhere around campus. C) The stern voices of teachers are heard across quads as they issue JUG to moronic students for the ...
While technically a dog whistle is an ultrasonic transducer that converts mechanical energy in the form of air pressure into ultrasonic sound waves, the term is more apt to be used to refer to piezoelectric transducers that convert electrical energy into sound. Piezoelectric crystals have the property of changing size when a voltage is applied, thus applying an alternating current (AC) across them causes them to oscillate at very high frequencies, thus producing very high frequency sound waves. The location at which a transducer focuses the sound can be determined by the active transducer area and shape, the ultrasound frequency, and the sound velocity of the propagation medium. The example shows the sound fields of an unfocused and a focusing ultrasonic transducer in water. 3. 1. 5. DETECTOR
Since piezoelectric crystals generate a voltage when force is applied to them, the same crystal can be used as an ultrasonic detector. Some systems use separate transmitter and receiver components while others combine both in a single piezoelectric transceiver. Alternative methods for creating and detecting ultrasound include magnetostriction and capacitive actuation. 3. 1. 6. SPEED OF SOUND The speed which sound travels depends on the medium which it passes through. In general, the speed of sound is proportional (the square root of the ratio) to the stiffness of the medium and its density. This is a fundamental property of the medium. Physical properties and the speed of sound change with the conditions in the environment.
The speed of sound in the air depends on the temperature. In the air speed is approximately 345 m/s, in water 1500 m/s and in a bar of steel 5000 m/s. A common use of ultrasound is for range finding. This use is also called sonar. Sonar works similarly to radar. An ultrasonic pulse is generated in a particular direction. If there is an object in the way of this pulse, the pulse is reflected back to the sender as an echo and is detected. Measuring the difference in time between the pulse transmitted and the echo received. it is possible to determine how far away the object is. Bats use a variety of ultrasonic ranging (echolocation) to detect their prey. 3. 1. 7. FEATURES Provides precise, non-contact distance measurements within a 2 cm to 3 m range • Simple pulse in/pulse out communication • Burst indicator LED shows measurement in progress • 20 mA power consumption • Narrow acceptance angle • 3-pin header makes it easy to connect using a servo extension cable, no soldering required 3. 1. 8. SPECIFICATIONS • Power requirements: +5 VDC • Communication: Positive TTL pulse • Dimensions: 0. 81 x 1. 8 x 0. 6 in (22 x 46 x 16 mm) • Operating temp range: +32 to +158 °F (0 to +70 °C) 3. 2. MICROCONTROLLER ATMEL 89S52 3. 2. 1. ATMEL 89S52 The AT89S52 is a low-power, high-performance CMOS 8-bit microcontroller with 8K bytes of in-system programmable Flash memory.
The device is manufactured using Atmel’s high-density nonvolatile memory technology and is compatible with the industry standard 80C51 instruction set and pin out. The on-chip Flash allows the program memory to be reprogrammed in-system or by a conventional nonvolatile memory programmer. By combining a versatile 8-bit CPU with in-system programmable Flash on a monolithic chip, the Atmel AT89S52 is a powerful microcontroller which provides a highly-flexible and cost-effective solution to many embedded control applications. The AT89S52 provides the following standard features: 8K bytes of Flash, 256 bytes of RAM, 32 I/O lines, Watchdog timer, two data pointers, three 16-bit timer/counters, a six-vector two-level interrupt rchitecture, a full duplex serial port, on-chip oscillator, and clock circuitry. In addition, the AT89S52 is designed with static logic for operation down to zero frequency and supports two software selectable power saving modes. The Idle Mode stops the CPU while allowing the RAM, timer/counters, serial port, and interrupt system to continue functioning. The Power-down mode saves the RAM con-tents but freezes the oscillator, disabling all other chip functions until the next interrupt or hardware reset. 3. 2. 2. PIN CONFIGURATION [pic] Figure 3. 2. 2. Pin configuration 3. 2. 3. BLOCK DIAGRAM [pic] Figure 3. 2. 3. Block Diagram 3. 2. 4 PROGRAMMING THE FLASH – PARALLEL MODE
The AT89S52 is shipped with the on-chip Flash memory array ready to be programmed. The programming interface needs a high-voltage (12-volt) program enable signal and is compatible with conventional third-party Flash or EPROM programmers. The AT89S52 code memory array is programmed byte-by-byte. Programming Algorithm: Before programming the AT89S52, the address, data, and control signals should be set up according to the “Flash Programming Modes” (To program the AT89S52, take the following steps: 1. Input the desired memory location on the address lines. 2. Input the appropriate data byte on the data lines. 3. Activate the correct combination of control signals. 4. Raise EA/VPP to 12V. 5.
Pulse ALE/PROG once to program a byte in the Flash array or the lock bits. The byte-write cycle is self-timed and typically takes no more than 50 ? s. Repeat steps 1 through 5, changing the address and data for the entire array or until the end of the object file is reached. 3. 2. 5. DATA POLLING The AT89S52 features Data Polling to indicate the end of a byte write cycle. During a write cycle, an attempted read of the last byte written will result in the complement of the written data on P0. 7. Once the write cycle has been completed, true data is valid on all outputs, and the next cycle may begin. Data Polling may begin any time after a write cycle has been initiated. 3. 2. 6. READY/BUSY
The progress of byte programming can also be monitored by the RDY/BSY output signal. P3. 0 is pulled low after ALE goes high during programming to indicate BUSY. P3. 0 is pulled high again when programming is done to indicate ready. 3. 2. 7. PROGRAM VERIFY If lock bits LB1 and LB2 have not been programmed, the programmed code data can be read back via the address and data lines for verification. The status of the individual lock bits can be verified directly by reading them back. 3. 2. 8. READING THE SIGNATURE BYTES The signature bytes are read by the same procedure as a nor-mal verification of locations 000H, 100H, and 200H, except that P3. 6 and P3. must be pulled to a logic low. The values returned are as follows. (000H) = 1EH indicates manufactured by Atmel (100H) = 52H indicates AT89S52 (200H) = 06H 3. 2. 9. CHIP ERASE In the parallel programming mode, a chip erase operation is initiated by using the proper combination of control signals and by pulsing ALE/PROG low for a duration of 200 ns – 500 ns. In the serial programming mode, a chip erase operation is initiated by issuing the Chip Erase instruction. In this mode, chip erase is self-timed and takes about 500 ms. During chip erase, a serial read from any address location will return 00H at the data output. 3. 3. MOTOR DRIVER L298 3. 3. 1. H BRIDGE
The term H bridge is derived from the typical graphical representation of such a circuit. An H bridge is built with four switches (solid-state or mechanical).
When the switches S1 and S4 (according to the first figure) are closed (and S2 and S3 are open) a positive voltage will be applied across the motor. By opening S1 and S4 switches and closing S2 and S3 switches, this voltage is reversed, allowing reverse operation of the motor. Using the nomenclature above, the switches S1 and S2 should never be closed at the same time, as this would cause a short circuit on the input voltage source. The same applies to the switches S3 and S4. This condition is known as shoot-through. [pic] Figure 3. 3. 1. H Bridge L298 3. 3. 2. CONSTRUCTION
A solid-state H bridge is typically constructed using opposite polarity devices, such as PNP BJTs or P-channel MOSFETs connected to the high voltage bus and NPN BJTs or N-channel MOSFETs connected to the low voltage bus. The most efficient MOSFET designs use N-channel MOSFETs on both the high side and low side because they typically have a third of the ON resistance of P-channel MOSFETs. This requires a more complex design since the gates of the high side MOSFETs must be driven positive with respect to the DC supply rail. However, many integrated circuit MOSFET drivers include a charge pump within the device to achieve this. Alternatively, a switched-mode DC–DC converter can be used to provide isolated (‘floating’) supplies to the gate drive circuitry.
A multiple-output fly back converter is well-suited to this application. Another method for driving MOSFET-bridges is the use of a specialised transformer known as a GDT (Gate Drive Transformer), which gives the isolated outputs for driving the upper FETs gates. The transformer core is usually a ferrite torrid, with 1:1 or 4:9 winding ratio. However, this method can only be used with high frequency signals. The design of the transformer is also very important, as the leakage inductance should be minimized, or cross conduction may occur. The outputs of the transformer also need to be usually clamped by Zener diodes, because high voltage spikes could destroy the MOSFET gates.
A further variation is the half-controlled bridge, where one of the high- and low-side switching devices (on opposite sides of the bridge) are replaced with diodes. This eliminates the shoot-through failure mode, and is commonly used to drive variable/switched reluctance machines and actuators where bi-directional current flow is not required. A “double pole double throw” relay can generally achieve the same electrical functionality as an H bridge (considering the usual function of the device).
An H bridge would be preferable to the relay where a smaller physical size, high speed switching, or low driving voltage is needed, or where the wearing out of mechanical parts is undesirable.
There are many commercially available inexpensive single and dual H-bridge packages, and L293x series are the most common ones. Few packages have built-in fly back diodes for back EMF protection. 3. 3. 3. OPERATION The H-bridge arrangement is generally used to reverse the polarity of the motor, but can also be used to ‘brake’ the motor, where the motor comes to a sudden stop, as the motor’s terminals are shorted, or to let the motor ‘free run’ to a stop, as the motor is effectively disconnected from the circuit. The following table summarises operation, with S1-S4 corresponding to the diagram below. [pic] Figure 3. 3. 3. Circuit For Operation 3. 3. 4. FEATURES • Delivers up to 5 A continuous 6 A peak current.
Optimized for DC motor management applications • Operates at supply voltages up to 40 V. Very low RDS ON ; typ. 200 m? @ 25 °C per switch • Output full short circuit protected • Over temperature protection with hysteresis and diagnosis • Short circuit and open load diagnosis with open drain error flag • Under voltage lockout • CMOS/TTL compatible inputs with hysteresis • No crossover current • Internal freewheeling diodes • Wide temperature range; ? 40 °C < Tj < 150 °C 3. 3. 5. APPLICATION This Breakout-Module is DC motor driver using L298. This is most commonly used driver circuit in hobby robots. This can control 2 DC Motors, their direction using control lines and their speed using PWM. 3. 4.
STEP DOWN TRANSFORMER A transformer is a static electrical device that transfers energy by inductive coupling between its winding circuits. A varying current in the primary winding creates a varying magnetic flux in the transformer’s core and thus a varying magnetic flux through the secondary winding. This varying magnetic flux induces a varying electromotive force (EMF), or “voltage”, in the secondary winding. Transformers range in size from thumbnail-sized units hidden inside microphones to units weighing hundreds of tons used in power grid applications. A wide range of transformer designs are used in electronic and electric power applications.
Transformers are essential for the transmission, distribution, and utilization of electrical energy. 3. 4. 1. FEATURES OF STEP DOWN TRANSFORMER • 120 Volts, 50/60HZ Reduced To 12V or 24V With • Outputs From 100VA to 1000VA • Automatic Resetting Protector in Primary Windings • Potted Core and Coil for Cool, Quiet Operation • Wiring Provided for Three Output Voltages • Low Voltage Re-Settable Supplementary Protector. • One Protector for Each Output Circuit • Supplied Protectors Service Both 12V AND 24V • All GP series transformer inputs are suitable for connection • to dimmers specifically designed for use wih magnetic or • inductive circuits. 3. 5. RECTIFIER
A rectifier is an electrical device that converts alternating current (AC), which periodically reverses direction, to direct current (DC), which flows in only one direction. The process is known as rectification. Physically, rectifiers take a number offorms, including vacuum tube diodes, mercury-arc valves, solid state diodes, silicon-controlled rectifiers and other silicon-based semiconductor switches. Historically, even synchronous electromechanical switches and motors have been used. Early radio receivers, called crystal radios, used a “cat’s whisker” of fine wire pressing on a crystal of galena (lead sulfide) to serve as a point-contact rectifier or “crystal detector”.
Rectifiers have many uses, but are often found serving as components of DC power supplies and high-voltage direct current power transmission systems. Rectification may serve in roles other than to generate direct current for use as a source of power. As noted, detectors of radio signals serve as rectifiers. In gas heating systems flame rectification is used to detect presence of flame. The simple process of rectification produces a type of DC characterized by pulsating voltages and currents (although still unidirectional).
Depending upon the type of end-use, this type of DC current may then be further modified into the type of relatively constant voltage DC characteristically produced by such sources as batteries and solar cells. 3. 6. REGULATOR
Battery balancing and battery redistribution refer to techniques that maximize a battery’s capacity to make all of its energy available for use and increase the battery’s lifetime. A battery balancer or battery regulator is a device in a battery pack that performs battery balancing. [1] Balancers are often found in Lithium ion battery packs for cell phones and laptop computers. They can also be found inbattery electric vehicle battery packs. Typically, the individual cells in a battery have somewhat different capacities and may be at different levels of state of charge (SOC).
Without redistribution, discharging must stop when the cell with the lowest capacity is empty (even though other cells are still not empty); this limits the energy that can be taken from and returned to the battery.
Without balancing, the cell of smallest capacity is a “weak point”, it can be easily overcharged or over-discharged while cells with higher capacity undergo only partial cycle. For the higher capacity cells to undergo full charge/discharge cycle of the largest amplitude, balancer should “protect” the weaker cells; so that in a balanced battery, the cell with the largest capacity can be filled without overcharging any other (i. e. weaker, smaller) cell, and it can be emptied without over-discharging any other cell. Battery balancing is done by transferring energy from or to individual cells, until the SOC of the cell with the lowest capacity is equal to the battery’s SOC. Battery redistribution is sometimes distinguished from battery alancing by saying the latter stops at matching the cell’s state of charge(SOC) only at one point (usually 100% SOC), so that the battery’s capacity is only limited by the capacity of its weakest cell 3. 7. FUSE In electronics and electrical engineering, a fuse (from the French fuser, Italian fuso, “spindle”[1]) is a type of low resistance resistor that acts as a sacrificial device to provide overcurrent protection, of either the load or source circuit. Its essential component is a metal wire or strip that melts when too much current flows, which interrupts the circuit in which it is connected. Short circuit, overloading, mismatched loads or device failure are the prime reasons for excessive current.
A fuse interrupts excessive current (blows) so that further damage by overheating or fire is prevented. Wiring regulations often define a maximum fuse current rating for particular circuits. Overcurrent protection devices are essential in electrical systems to limit threats to human life and property damage. Fuses are selected to allow passage of normal current plus a marginal percentage and to allow excessive current only for short periods. Slow blow fuses are designed to allow harmless short term higher currents but still clear on a sustained overload. Fuses are manufactured in a wide range of current and voltage ratings and are widely used to protect wiring systems and electrical equipment.
Self-resetting fuses automatically restore the circuit after the overload has cleared; these are useful, for example, in aerospace or nuclear applications where fuse replacement is impossible. CHAPTER 4 NEURAL NETWORK 4. 1. PROBLEM FORMULATION IN NEURAL NETWORK 4. 1. 1. PRELIMINARIES The goal of a blind spot surveillance system is to assist the driver when changing lanes and avoid possibly dangerous situations. The blind spot zone ranges from 3 m behind the car to the side mirrors and 3 m laterally. A warning signal shall be emitted when a car occurs within this zone during a lane change. To ensure applicability, some preconditions are necessary.
For an optimal performance, a maximal reaction time of 300 ms is desired and the overall detection time should not exceed 1500 ms. A low false-alarm-rate is also required since too many missed or unnecessary warnings corrupt the driver’s faith in the system’s reliability. The operating range must be designed to detect blind spot alerts to a speed difference between the host and traffic from 0 to 30 km=h. 4. 1. 2. SYSTEM SETUP The host vehicle is equipped with ultrasonic sensors equally positioned at its front and its back side. To detect overtaking vehicles, the approach at hand evaluates the measurements of two sensors on each side of the host. All other sensors are not used for blind spot detection.
The aperture of the two rear sensors is approximately 75, while the aperture of the front sensors is set to 45. This enables sharp measurements with the front sensor in order to detect incoming traffic from the front or outgoing traffic from the back. In case of traffic residing within the blind spot zone, the driver is notified by illuminating a red light in its side mirror. 4. 1. 3. NEURAL NETWORKS Since the curve data base is calculated over a lattice of orthogonal distance and relative velocity, it is impossible to detect a unique fitting function in most cases. There are several functions with similar deviation values. An artificial neural network is able to tackle this problem and refine the decision process.
In this paper a feed-forward neural network, which means that there are no cycles within, containing twenty neurons within the hidden layer is used. The training was realized by supervised learning using the recorded data from different test drives as input values for the Levenberg-Marquardt-Algorithm. This input values are the sample mean and covariance of all sensors in the current moving window Wm, the results of the curve-fitting-process represented by a state variable indicating whether an approaching car, a stationary object, constant distance or none of those cases has been detected and the sum of the detected states covering the last second of measurements.
The approach at hand. The neural network’s binary output value is 1 if a possibly dangerous state has been detected, 0 otherwise. A warning is emitted, if two of the last three output values are nonzero. [pic] Figure 4. 1. neural network There are several ways to realize this neural network approach. One possibility is to train different networks for each environment namely inner city, interurban and motorway traffic. Another idea is to design some kind of one-size-fits-all network, whose training set contains a mix of test drives in different environments. The work at hand illustrates both possibilities demonstrating three neural networks Nauto Ncity and Nmix.
The training set of Nauto contains three motorway files with 45 km driven distances and 88 overtaking maneuvers, Ncity was trained using two inner city files with 10 km driven distance and 46 overtaking maneuvers and the underlying training data of Nmix is a combination of these motorway and inner city test drives in a ratio of 3:2 with 55 km driven distance and 134 overtaking maneuvers. In this first attempt interurban drives have been left out of the training files since the chosen networks are expected to cover that cases at a satisfying. [pic] Figure 4. 2. Detection rates without host speed limits vehicle was equipped with one laser sensor on each side and four color cameras mounted on top of the car to generate a 360_ view of the environment. In order to ensure meaningful results, differing types of target vehicles like cars, motorbikes or trucks and different road environments like inner city, interurban or motorway drives had to be considered.
After more than 2000 km of test drives, the data base contains over 3000 test cases for qualitative and quantitative evaluation. The three networks Nauto;Ncity and Nmix have been applied to a collection of test files containing approximately 356 km driven on motorways and addionally about 20 km and 32 km driven in city respectively interurban traffic. According to the requirements stated in the False-alarm-rate for all networks and different host speed intervals desired maxmial reaction time shall not exeed 300 ms. Hence detection intevals of 0:3 s, 0:6 s, 1:5 sand the overall detection rate without any time limit are statistically evaluated. [pic] Figure 4. 3.
Detection rates within 25 km=h to 50 km=h (moderate speed) The performance of the networks is compared in three settings with different host speed intervals to illustrate their particular strengths. the detection rates for the different speed-intervals. The first has no limitations concerning the host vehicle’s speed to illustrate the overall performance of all networks. the results in the second interval (moderate-speedinterval) ranging from 25 km=h to 50 km=h to evaluate the inner city efficiency. Finally the minimal host speed of the third interval (high-speed-interval, set to 70 km=h . the false-alarm-rate of all networks relative to the total number of emitted warnings. The results demonstrated show that the networks Nauto and Ncity achieve promising detection rates for the particular driving situations they have been trained for.
As expected Ncity performs best in terms of moderate velocities where a slightly elevated reaction time is acceptable achieving an overall detection rate of 96:3% and even 84% within 0:6 s. Analogously, Nauto provides satisfying detection rates in terms of high velocities detecting all vehicles and even 93:1% within 0:3 s. The one-size-fitsall network Nmix provides low false-alarm-rates in the overall and high-speed setting in exchange for a slightly elevated reaction time but still achieving overall detection rates of at least 96:3% in every setting [pic]. Figure 4. 4. Detection rates within high-speed-interval (over 70 km=h) As a start the results of the neural network approach at hand show promise.
Since every network has its strengths in particular situations, there are several possibilities for future investigations. Although the one-size-fits-all network provides a solid overall performance the training of different networks for several situations is prefered since the specialized networks provide even better detection rates for their particular strengths within less reaction time. Since it is possible to detect the actual traffic situation via odometry and curvature information, a deeper analysis of this approach is intended. Another aspect demanding further investigations is a neural network trained for rain weather conditions including wet roads and splash water.
