(1st) Project - SELF-DRIVING VEHICLE

SELF-DRIVING VEHICLE

A Focused Strategic Effort towards the Autonomous Vehicle Industry


Name : Shubham Tiwary, 2nd year, CSE, GCETTB


INTRODUCTION

 

Every year, millions of lives are lost to traffic crashes around the world, and in India the number of tragedies is growing. A common element of these crashes is that 94% involve human error. Driving is not as safe or as easy as it should be, while distracted driving is on the rise. We believe our technology could save thousands of lives now lost to traffic crashes every year.

 

Fully self-driving vehicles are indeed a solution that will succeed in its promise and gain public acceptance only if they are safe. That’s why Engineers across the globe have been investing in safety and building the processes that give us the confidence that the self-driving vehicles can serve the public’s need for safer transportation and better mobility.

 

 

 

ABOUT THE PROJECT

 

Self-driving vehicles are a maturing technology with the potential to reshape mobility by enhancing the safety, accessibility, efficiency, and convenience of automotive transportation. Safety-critical tasks that must be executed by a self-driving vehicle include planning of motions through a dynamic environment shared with other vehicles and pedestrians, and their robust executions via feedback control. The objective of this paper is to survey the current state of the art on planning and control algorithms with particular regard to the urban setting. A selection of proposed techniques is reviewed along with a discussion of their effectiveness. The surveyed approaches differ in the vehicle mobility model used, in assumptions on the structure of the environment, and in computational requirements. The side-by-side comparison presented in this survey helps to gain insight into the strengths and limitations of the reviewed approaches and assists with system level design choices.

With this project I have written and implemented a controller for the CARLA Simulator. I have integrated vehicle modeling and controller design into a complete vehicle control system and developed a working simulation with a python-based vehicle autonomy agent (CARLA), tuned a control system for tracking performance on a complex path both longitudinally and laterally.

 

Though the process a little messy but if we break down the steps of Driving, automating it will have a clearer path for its execution: The task of driving can be further broken down in these steps:

 

1.     Perceiving the Environment

2.     Planning the route from A to B

a.     Long Term Planning – Start and End Point

b.    Short Term Planning – Overtaking, Lane Change

3.     Controlling the vehicle

a.     Lateral Control – Steering

b.    Longitudinal Control – Braking, Accelerating

c.     Object and Event Detection and Response – Detection, Response

 

 

 

 

HARDWARE EQUIPMENTS

 

Perceiving and knowing about the environment plays a very important role in deciding and planning how to drive and what actions to take. Humans perceive about the environment through their sense organs for manual driving as all the actions are taken by them in the first place, but computer have to be loaded with certain specific hardware to perceive the environment: the hardware components are:

 

 

SOFTWARE EQUIPMENTS

 

Since errors are very common in software development and testing, it would be a bad idea to test our software directly on vehicles which would be cost ineffective as well as potentially dangerous for people. Well it has a solution, we have an unreal engine, an open-source software for autonomous driving research. Software for simulation is developed from the ground up to support development, training, and validation of autonomous urban driving systems. In addition to open-source code and protocols, CARLA(the software used in this project for simulation) provides open digital assets (urban layouts, buildings, vehicles) that were created for this purpose and can be used freely. The simulation platform supports flexible specification of sensor suites and environmental conditions. We use CARLA to study the performance of three approaches to autonomous driving: a classic modular pipeline, an end-to-end model trained via imitation learning, and an end-to-end model trained via reinforcement learning. The approaches are evaluated in controlled scenarios of increasing difficulty, and their performance is examined via metrics provided by CARLA, illustrating the platform’s utility for autonomous driving research.

The installation and operational methods are in the references section of this document.

 



UPGRADES TO THE SOFTWARE (FUTURE PLAN)

 

The Controller Software whose implementation you just saw is just an example of Level 2 Autonomy. The Software can be improved so can its Level of Autonomy.

Improvements to the model can be:

1.     Function for monitoring the driving environment.

2.     Function for detection, recognition and classification of objects on road and happening events.

3.     Function for executing appropriate response to object and events.

4.     Function to read Traffic Lights or Signs.

5.     Function for Fallback in case of any accident


Images :


PDF link : Link

Comments

Anonymous said…
Excellent concept bro