Xavier Institute of Engineering

Department of Computer Engineering

(Academic Year: 2020-2021 Semester: VII)

Class: BE

Subject: AISC

Experiment No. 8
Case Study of an Application: Simulation of Self driving car.
Title

Lab Outcome LO4

Achieved

Date of Performance //2020

Date of submission //2020
Roll No. 66
Name of the Student Aman Sharma

Rubrics used for Laboratory Evaluation:

Below
Good
Expectation Average

Knowledge (4) 2 3 4

Performance (5) 2 3 5

Content & Neatness of Documentation(3) 1 2 3

Punctuality & Submission on Time (3) 1 2 3

Marks Obtained: _________________________

Signature of Faculty: _______________________

 Experiment No.8

Aim: Case Study of an Application: Simulation of Self Driving Car.

Theory:
Self-driving cars, have rapidly become one of the most transformative technologies to emerge.
Fueled by Deep Learning algorithms, they are continuously driving our society forward, and
creating new opportunities in the mobility sector. Over the past decade, interest has increased
in self-driving cars.
Self-driving simulations provide multiple signals simultaneously, such as camera, LiDAR, radar
and more. A given object in the virtual environment will thus be detected by different sensors,
and these signals will validate each other to increase accuracy.
1. The race to autonomy:

 Real-time route optimization

 Increased lane capacity
 Reduced energy consumption

2. Driven by tech: There are several critical technologies behind safe and efficient
autonomous-vehicle operation—AI, safety and security, cameras, network infrastructure,
and the sensor technologies radar and lidar, or laser-light radar.
3. Artificial intelligence: AI is a major focus for autonomous-vehicle testing and development,
and the vehicles are applying AI—a collection of discrete technologies—in new and
innovative ways.
4. Safety and security: Autonomous vehicles won’t gain widespread acceptance until the riding
public feels assured of their safety and security, not only of passengers but also other
vehicles and pedestrians.
5. Sensor technology and Keep Driving

Proposed System:

 Homogenization and decoupling:

Homogenization comes from the fact that all digital information assumes the same form.
During the ongoing evolution of the digital era, certain industry standards have been
developed on how to store digital information and in what type of format. This concept of
homogenization also applies to autonomous vehicles. In order for autonomous vehicles to
perceive their surroundings, they have to use different techniques each with their own
accompanying digital information (e.g. radar, GPS, motion sensors and computer vision).

 Connectivity:

Connectivity means that users of a certain digital technology can connect easily with other
users, other applications or even other enterprises. In the case of autonomous vehicles, it is
 essential for them to connect with other 'devices' in order to function most effectively.
Autonomous vehicles are equipped with communication systems which allow them to
communicate with other autonomous vehicles and roadside units to provide them, amongst
other things, with information about road work or traffic congestion.

 Reprogrammable:

Another characteristic of autonomous vehicles is that the core product will have a greater
emphasis on the software and its possibilities, instead of the chassis and its engine. This is
because autonomous vehicles have software systems that drive the vehicle meaning that
updates through reprogramming or editing the software can enhance the benefits of the
owner (e.g. update in better distinguishing blind person vs. non-blind person so that the
vehicle will take extra caution when approaching a blind person)
One potential benefit of self-driving cars is that they could be safer than human drivers who
are prone to distraction, fatigue and emotional decisions that lead to mistakes. But to
ensure safety, autonomous vehicles must evaluate and respond to the driving environment
without fail. Given the innumerable situations that a car might encounter on the road, an
autonomous driving system requires hundreds of millions of miles worth of test drives
under challenging conditions to demonstrate reliability.
Block diagram:

 High-level system architecture:

 Model Prediction:
Structures used:
Data collection:
We used Udacity's self-riding car simulator Collecting facts. This simulator is fashioned in Unity
And became employed by using Eudicity for self-using Nanodegree program however changed
into currently open. This explains what NVIDIA did Simulation.
The schooling mode is used to collect information and Autonomous mode is used to peer the
model. Additionally, there are two types of tracks Simulator - lake music and jungle song. The
lake song is quite small and smooth to address When the automobile compared with the jungle
track. The simulator captures Data when driving across the song using the automobile Left and
right keys to manipulate steerage angle Up and down arrows to manage the pace.
Udacity Simulation-Lake Track:

Udacity Simulator- Jungle track:

Design – flowchart:

Implementation:
 Data preprocessing:
The data we collect i.e. the captured image the model is preprocessed before being trained.
Whereas Preprocessing is cropped to extract images the sky and the front of the car. The
pictures are Then we are converting it from RGB to YUV and resizing the image. The input
size used by the model. is it done?
Because RGB is not the best mapping for the view an assumption. YUV has too much color
space, efficient coding and reduces bandwidth than RGB can capture.
 After selecting the last set of frames, the data is Enhanced by adding artificial shifts and
curves How to teach a network to overcome the poor
Position or orientation. While growing up, we Randomly select right, left or center images,
randomly flip left / right images and adjust

 Deep learning model:

The intensive learning model we've developed is predicated on Research is done by NVIDIA
for his or her autonomous Vehicle. The model includes the subsequent Critical layers.
i. Conversational layer: The convolution layer applies the convolution The of
Convolution function filter on the input image and produces a 3D production of
activation neurons.
ii. Max-pooling layer: Help reduce pooling/down sampling layer Number of
parameters/loads within the network and Helps reduce training time without losing
time Knowledge of any specific feature of the image.
iii. Thick layer: Is similar to a dense layer or a totally attached layer Common neural
network layer that contains all the neurons during this layer are connected to every
of the neurons from the previous layer.

NVIDIA Model in Simulation of Self-Driving Car:

Conclusion: Hence we have performed case study on simulation of Self driving car.