AI Based Driverless Car Using High Computational Embedded System

Project Summary The aim of this project is to design a system having multiple features related to AI-based obstacle/object detection, path detection, line tracking etc. Driverless car problem set consist of several sub domains i.e. machine learning, Artificial intelligence, i

Project Title

AI Based Driverless Car Using High Computational Embedded System

Project Area of Specialization

Artificial Intelligence

Project Summary

Project Summary

The aim of this project is to design a system having multiple features related to AI-based obstacle/object detection, path detection, line tracking etc. Driverless car problem set consist of several sub domains i.e. machine learning, Artificial intelligence, image processing etc. The main objective of the project will be achieved by establishing a system which acquires the images of the path in training phase and then predicting path accordingly. After training our vehicle, it will acquire images of its path using its camera and then it will recognize the obstacles/objects automatically and will detect and follow the path using Artificial intelligence. The images will be acquired through cameras mounted on the vehicle and Python/MATLAB will be used for Data Processing. After path detection the system will issue desired commands to Vehicle such as forward, left, right, stop etc. according to the given path. Some of the main key features that arose the need of this type of product are; less safer roads due to human errors, less productivity due to time consuming nature of currently used manual vehicles, higher cost due to accidents etc., less efficiency of manual vehicles as they includes human errors and last but not the least currently used vehicles are less environment friendly etc. With rapid economic development, intelligent vehicles are in urgent need. Along with the sustained and rapid growth of car ownership, almost every country is facing severe traffic congestion, road safety and environmental pollution problems. In the meanwhile, the number of fatal traffic accidents is increasing each year and most of them are caused by human operating errors. In fact, a study by the National Highway Traffic Safety Administration (NHTSA) revealed that 94% of accidents were caused by the drivers themselves. With the continued growth of car ownership, the number of fatal traffic accidents is expected to grow. Also currently used vehicles has less mobility options as they are not suitable for elderly, young and disabled users. Also by using manual vehicles, congestion rate is very much higher. Relying on advanced AI techniques, we can solve the aforementioned problems.

Project Objectives

Project Objectives

Some of the main objectives of our project are as follow.

• Machine Learning

Main objective of this project is to learn and perform machine learning on our self-driving robot car. With the help of machine learning, our self-driving robot car will be able to react according to the condition. In machine learning we will train our robot using different algorithms.

• Artificial Intelligence

Another objective of our project is to perform artificial intelligence. Artificial Intelligence has created a huge impact on the world’s economy and will continue to do so since we’re aiding its growth by producing an immeasurable amount of data. By using AI on our self-driving robot, it will be able to make best possible decisions by itself.

• Image Processing

Image processing is also an important objective of our project. Here in our project image processing is dealing with capturing real time images/videos and then converting them into frames. Then noise will be removed from those frames and object/path tracking will be performed using different image processing algorithms.

• Object Recognition

One of the core objective of our project is to recognize different objects on its path. First of all our self-driving robot car will identify/recognize its path and then while moving on its path it will detect objects on its way. For object detection and path recognition we will be using several high speed algorithms.

• Robot Building

Without building our self-driving robot car our project will be incomplete so it’s an important objective of our project to learn and build our robot. In robot car building there can be several phases such as components selection etc. By making our self-driving robot car by our own self we will have a better understanding of our product.

• Robot Movement

Learning and performing movement of our robot car is another important objective. Its movement will basically depend upon AI based decision making and object detection algorithms. Our self-driving robot car will be able to decide its movement according to condition.

Project Implementation Method

Project Implementation

As we know Artificial Intelligence has created a huge impact on the world’s economy and will continue to do so since we’re aiding its growth by producing an immeasurable amount of data. Thanks to the advancement in Artificial Intelligence Algorithms we can deal with such humungous data.

Generally, an algorithm takes some input and uses mathematics and logic to produce the output. In stark contrast, an Artificial Intelligence Algorithm takes a combination of both – inputs and outputs simultaneously in order to “learn” the data and produce outputs when given new inputs. AI algorithm will perform image processing which is running in Jetson Nano. Algorithm will perform image processing, which includes removing noise from image and then it will turn colour image into grey scale and then apply Morphological Operations and a Canny (Edge Detection) filter to detect the edges of path. After that each frame will me cropped to the interested region (path). Lastly, it will combine Line segments into two lanes and will move on path.

In our project, AI based car the car is equipped with a CSI camera and Jetson Nano for image processing. The camera will take real time input though camera and feed that input to Jetson Nano, then Jetson Nano will run AI algorithm on the input and send the instruction to Arduino through its data pins. Arduino will then send instructions to motors. If camera detects obstacle in the way of car then it will be processed by Jetson Nano and then Jetson Nano will give instructions to Arduino to stop the motors. And when the obstacle is removed from the path, it will start moving towards its destination.

Benefits of the Project

Project Benefits

Development of sustainable AI robot having capability of decision making. Artificial Intelligence is the ability of a computer program to learn and think. Artificial intelligence (AI) is pushing the boundaries of machine-enabled functionalities. This cutting-edge technology facilitates machines to act with a degree of autonomy, resulting in effective execution of iterative tasks.

Artificial Intelligence (AI) facilitates us in creation of next-generation environment that will enable us to developed new technology with less time and effort. AI provides organisations with the luxury of freeing up resources for higher-level tasks.

AI operates 2x7 without interruption or breaks and has no downtime. Furthermore, AI has mass market potential, it can be deployed across industries. AI can reduce human errors, the phrase “human error” was born because humans make mistakes from time to time. Computers, however, do not make these mistakes if they are programmed properly. This is one of the biggest advantages of Artificial intelligence that we can overcome many risky limitations of humans by developing an AI Robot which in turn can do the risky things for us.

Artificial Intelligence can assist the drivers in protecting against unexpected risks. Now, we can see there are many AI-enabled systems or sensors which give alarms to the driver to avoid risks/ accidents. In 2015, more than 38,000 people died in motor vehicle accidents per the National Safety Council. Of these accidents, 26 percent led to deaths that were caused by distracted driving.

AI based car seemed future but now the future is here. AI car that drivers itself and reaches at its destination with safety. No human input is required while driving. The car will automatically stop, if it detects any obstacle in the way. And when the obstacle is removed it will start moving towards its destination. This will help to avoid accidents and also AI car will follow all the traffic rules which will result in no traffic jam, no accidents etc. These risks will greatly be reduced, if not eliminated, through the introduction and widespread adoption of level 5 autonomous vehicles. Today, we are already starting to see the benefits with semi-autonomous vehicles such as a Tesla.

Technical Details of Final Deliverable

Project Deliverables:

• Self-Driving on basis of Artificial Intelligence

Artificial Intelligence will be used for driving instead of a human guiding the vehicle to move. Self-driving car is a driverless car which takes input from Jetson Nano instead of human. It uses AI to make decision and all the decisions are according to the condition faced by car. Our car uses AI algorithms that will be executed on Jetson Nano. Jetson Nano will get real-time input through camera and process it using AI algorithm which will then result in a perfect decision for the car.

• Decision Making using Embedded GPU

One of our main goal to be achieved is to perform decision making using high computational embedded system. Our Self-Driving car is using a high speed embedded system for decision making. The high computational embedded system that is used in our project is Jetson Nano. Jetson Nano itself is a GPU (Graphical Processing Unit). As GPU’s are designed for fast processing of graphical data so it is a fast and reliable system which process all the information and make a decision at run time. NVIDIA Jetson Nano Developer Kit is a small, powerful computer that runs multiple neural networks in parallel for applications like image classification, object detection, segmentation, and speech processing. All in an easy-to-use platform that runs in as little as 5 watts.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Security

Other Industries

Transportation

Core Technology

Artificial Intelligence(AI)

Other Technologies

Robotics

Sustainable Development Goals

Industry, Innovation and Infrastructure, Partnerships to achieve the Goal

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com