Autonomous Cars

[Virtual Presenter] Hello my name is Ahmed Hasan and today we will be talking about Autonomous Cars and how they work I hope you discover a new thing today.

[Audio] After you understand this unit, you will be able to Demonstrate an autonomous car and You will be able to discuss the problems of using autonomous cars.

[Audio] Table of contents IS autonomous car Machine learning and Artificial intelligence.

[Audio] The Key terms is that Artificial intelligence is a machine that has the ability to finish a human task that requires human intelligence And Machine learning is a type of algorithm that enable machines to learn from data and An autonomous car is a car that has the capability to receive data and then understand its environment without the help of humans.

[Audio] We shall learn about autonomous vehicles in this part. This is one of the uses for artificial intelligence (AI) and machine learning (ML) that comes to mind. We will discover how data collection and machine learning techniques are used to educate autonomous vehicles to drive on their own..

[Audio] Definition of autonomous car An autonomous car is a self-driving vehicle that can comprehend its surroundings and function without assistance from others. Using sensors, it can comprehend the environment. The information gathered by sensors is used to teach the car to abide by traffic laws and steer clear of obstacles..

[Audio] Also An average autonomous vehicle may include more than 20 sensors to comprehend its surroundings. For instance, the Tesla 3 contains one radar, twelve ultrasonic sensors, and eight cameras..

[Audio] Autonomous car main parts You will discover the essential components of an autonomous car in this section. But first, look up online the components that autonomous vehicles use to analyze and navigate their surroundings. The main parts of an autonomous car include sensors for data collection. a computer to receive, process and send signals. an ML algorithm that will be used to create and train an ML model. motors to move the car..

[Audio] Ml model definition machine learning model uses data from an unexplored dataset to uncover patterns or make choices. For instance, in the field of natural language processing, machine learning models are able to analyze and accurately identify the intention underlying previously unheard sentences or word combinations..

[Audio] the Sensors used to collect data Autonomous cars need to understand the surrounding environment to be able to navigate it. This is done using sensors. This includes cameras, radars, and ultrasonic sensors. The collected data is used to create an ML model which enable the car to identify objects around..

[Audio] Data processing A processing unit is required for autonomous vehicles in order to process data and manage movement. A specially constructed computer with a processing speed of several trillion operations per second was needed to build the actual autonomous vehicle.

[Audio] The AI processIt is An image classification model that can be utilized in an autonomous vehicle was made using the Al method. The following is a list of the four steps that you learned in earlier terms. 1 Data collection 2 Data representation 3 Training 4 Testing.

[Audio] Data Collection. Data collection is the process of acquiring and analyzing information on relevant variables in a predetermined, methodical way so that one can respond to specified research questions, test hypotheses, and assess results..

[Audio] Also The image consists of pixels. A pixel is the smallest unit of an image. Each pixel has 3 sub- pixel that controls the amount of red, green, and blue lights..

[Audio] AND As seen in the illustration, numbers can be used to represent how much light is contained in each pixel. The figure shows that a matrix of numbers can be used to represent the pixels. The value of a pixel in the matrix increases with pixel darkness. Three matrices are piled on top of one another in a colored image. Each matrix keeps track of a particular red, green, or blue value..

[Audio] and after that In a data structure that resembles matrices, as illustrated in the pictures, the gathered photos are kept..

[Audio] Data representation In data representation The data is prepared to create a dataset to train the ML model during the second step of the Al process. The amount of pixels in an image determines the size of the image. For instance, an image that was 64x64 in size would have 4096 pixels total..

[Audio] Image understanding and labelling AI will turn vectors into pixels to understand the meanings of it because ML only understands numbers that's machine language..

[Audio] Training The best ML algorithm used to classify images is the convolutional neural network (CNN). CNN algorithms are able to find unique shapes and patterns in the images..

[Audio] The process of training a CNN model is iterative, meaning that it is continually repeated in order to improve the model. The model's performance is a little underwhelming in the first edition. However, each iteration results in improved performance. Typically, training uses 70% to 80% of the data, whereas testing uses the remaining 20% to 30%..

[Audio] Tasting To determine how effective our model is, the ML model's performance must be evaluated. This is accomplished by calculating the proportion of accurate predictions to all predictions. That demonstrates the model's accuracy. Accuracy = Correct prediction dvided by Total number of predictions The testing dataset, which typically comprises 20% to 30% of the overall dataset, is used to test the model's performance..

[Audio] This is the end of our lesson I hope you learn something new See you in the second class.