0% found this document useful (0 votes)
301 views9 pages

Auto Drive PDF

The document discusses Drive.ai's approach to autonomous driving which involves training their self-driving cars using machine learning similar to how humans are trained. It provides details about their ground truth labeling process, sensor fusion algorithms, situation generation capabilities, and visualizations. Key features of Drive.ai's system include sensor integration, situation generation, computer vision algorithms for path detection and more.

Uploaded by

Chris Leyba
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
301 views9 pages

Auto Drive PDF

The document discusses Drive.ai's approach to autonomous driving which involves training their self-driving cars using machine learning similar to how humans are trained. It provides details about their ground truth labeling process, sensor fusion algorithms, situation generation capabilities, and visualizations. Key features of Drive.ai's system include sensor integration, situation generation, computer vision algorithms for path detection and more.

Uploaded by

Chris Leyba
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

3/12/2018 Auto drive

Phases One
Introduction
Center to Drive.ai's approach is utilizing profound learning no matter how you look at it in its self-sufficient
driving framework, which implies they're instructing their self-driving autos fairly like how you'd educate a
human. That includes giving a large group of cases of circumstances, articles and situations and afterward giving
the framework a chance to extrapolate how the guidelines it realizes there might apply to novel or sudden
encounters.

Key Feature
Ground-truth marking work process application to computerize naming, and apparatuses to contrast
recreation yield and ground truth
Sensor combination and following calculations, including Kalman channels, multiobject following
structure, location track task, and movement models
Driving situation age, including street, performer, and vehicle definition and situation perceptions
Sensor recreation for camera and radar, with protest records as yield
Computer vision calculations, including path identification and characterization, vehicle discovery, and
picture vehicle facilitate changes
Drive ai core feature- Self-Driving
Visualizations, including bird's-eye-see plots of sensor scope, discoveries, and tracks, and video overlays
for path markers and vehicle discovery

Look

file:///C:/Users/IK/Desktop/SE/201701249.html 1/9
3/12/2018 Auto drive

Attachment: 201701249-5_1_1.jpg

User Requirement
The user requirement specification meaning is a document usually used in SE(Software engineedring) that
specifies that user want from the software to be able to do.

Self-Driving
self-driving is very special in the modern revolution. In this step Autonomous cars are driven by distance
program or drives through internet connection.

Visualize

Visualization

· Autonomous cars engineering it’s a smart look at an impressive platform of viewing the new generation. In this
cars have required cameras to view 3600 so that it can drive safely.

file:///C:/Users/IK/Desktop/SE/201701249.html 2/9
3/12/2018 Auto drive

Attachment: 201701249-7_1_self-driving-car-sensor-areas.png

Tracking

In this kind of cars us their GPS signals from the satellites to keep in touch with the cars and combined with
readings from tachometer, altimeters and gyroscopes to provide more accurate positioning.

Visualize
Autonomous cars engineering it’s a smart look at an impressive platform of viewing the new generation. In this
cars have required cameras to view 3600 so that it can drive safely.

file:///C:/Users/IK/Desktop/SE/201701249.html 3/9
3/12/2018 Auto drive

Attachment: 201701249-11_1_self-driving-car-sensor-areas.png

Tracking
In this kind of cars us their GPS signals from the satellites to keep in touch with the cars and combined with
readings from tachometer, altimeters and gyroscopes to provide more accurate positioning.

Annotate video Using detection


Video cameras to detect the traffic lights, read road signs, keep track of the position of the vehicle.

Functional Requirement
A System Requirements Specification (SRS) (also known as a Software Requirements Specification) is a
document or set of documentation that describes the features and behavior of a system or software application.

Interior
Now a days autonomous cars are just getting a new auto designs and designers are already working on a new
generation of interiors.

Computer
Ever autonomous cars need a computer to control there ever moment and every system.

Backup System
file:///C:/Users/IK/Desktop/SE/201701249.html 4/9
3/12/2018 Auto drive

Autonomous cars that backup their computerized system with a remote human driver to operate that car instead
of a fallback driver.

Electric Batteries
Battery makers must rethink their technology if predictions for a wave of self-driving vehicles pan out,
according to one of the inventors of the lithium-ion battery.
Self-driving vehicles need more-durable batteries: Yoshino, EVs to account for 54% of new car sales by 2040:
BNEF forecast

Odometry sensors
In case of navigation Odometry sensors complement and improve GPS information. It helps the autonomous
cars to drive safely.

Ultrasonic
Ultrasonic sound for assisted parking and reflected sound waves detect distance to near objects. Some cars use
short-range radar instead.

V2V and V2I Communication


Self-driving vehicles integrated with vehicle-to-vehicle(V2V) communications and vehicle-to-
Infrastructure(V2I) communication.

EPS(Electric Power Steering)


Electric Power Steering Automation for Autonomous Driving Is really an important function. EPS is being fitted
to heavier vehicles such as SUVs and pickup trucks, fitted with electric motors which produce higher output
torque.

Park assist
park assist active when parking is needed and it uses an ultrasonic-based sensing system and electric power
assisted steering to positon the vehicle for parallel parking. It helps to Identified the nearest parking and if there
are no cars.

Attachment: 201701249-28_1_ford_active_park_assist_main630-1230-636x360-400x197.jpg

file:///C:/Users/IK/Desktop/SE/201701249.html 5/9
3/12/2018 Auto drive

ACC
Adaptive cruise control systems use as a radar to detect slower cars. The vehicles then adjust its speed to
maintain a safe distance from any vehicles ahead.

GPS
Global positioning system (GPS)signals from satellites are combined with readings from tachometer and
gyroscopes for more accurate positioning.

Sonar
Sonar-Use ultrasonic sensors to find the vehicle near in the parking area.

Radar
Radar sensors control the position of the vehicle nearby. Such sensors are already used in ACC system

Cameras
Video cameras detect traffic light, road signs and keep track of the position of other cars nearby

Lidar
Light detection and ranging sensors bounce pulse of light off the surroundings. There are analyses to identify
lane marks and the edges of the road.

Sensors
Sensors provides vehicle with the real time environment conditions

Stakeholders
Users
Developers
Insurances Industry
Professional drivers (Taxi, Limo, Bus, truck)
Energy & Fuel companies
Ride-hailing companies
Hotels and motels
Car manufactures
After market and auto repairs

Non-Functional Requirement

file:///C:/Users/IK/Desktop/SE/201701249.html 6/9
3/12/2018 Auto drive

In systems engineering and requirements engineering, a non-functional requirement (NFR) is a requirementthat


specifies criteria that can be used to judge the operation of a system, rather than specific behaviors. They are
contrasted with functional requirements that define specific behavior or functions.

Performance SLA
Performance SLA
Protocol choice(AMQP, NNTP, MQTT, etc.)
Message throughput
Traceability / Debug

Availability
Autonomous cars record system is required to have more accuracy on availability.

Safety
Autonomous care is more safer the robotic driving. Well navigation and improve road safety easy for AI to
drive. Few standards exist in connected vehicle.

Security
The risks is from cyber criminals. Have extensive experience developing strategic and practical approaches for
managing cybersecurity and privacy risks for a wide range of enterprises, including automotive companies

• Reliability (redundancy), Availability, Recoverability


In a dense urban setting, a driverless taxi can appear close to being summoned. In any case, not so in rural areas,
where individuals live miles apart. For some, delay and "loss of control" over their own mobility will build the
interest of owning their own particular vehicle.

Cost
On account of less difficult electric motors and lighter auto bodies, independent vehicles will be less expensive
to purchase and keep up than the present human-driven vehicles. A few evaluations convey the cost to $10K per
vehicle, a conspicuous difference with the present normal of $30K per vehicle.

Diversity of form and function


Autonomous vehicles will be accessible in a wide assortment of sizes and shapes. Buyers will drive interest for
specially crafted, reason constructed self-ruling vehicles whose frame is adjusted for a specific capacity.

Data privacy and integrity


In the database of an autonomous vehicle all kind of data is safe and secure. They care about their customers
privacy and integrity

Multiple sub-systems
file:///C:/Users/IK/Desktop/SE/201701249.html 7/9
3/12/2018 Auto drive

Inside of the autonomous vehicles are many sub-systems are working to do everything accordingly.

Communication board
long and short-rang traffic , broadcasting services, navigation and others are working so there nothing worst
happen

Domain requirements
Creation
Realization
Defeasibility
Unsafe
It will never work perfectly
Long and short-rang traffic

System architecture for the autonomously driving


vehicle
Attachment: 201701249-51_1_1.PNG

file:///C:/Users/IK/Desktop/SE/201701249.html 8/9
3/12/2018 Auto drive

Conclusion
Intelligent driving help capacities require a point by point comprehension of the vehicle's environment, of the
driving circumstance, and the activity standards and directions and in addition an adequate learning about the
material science of autos to satisfy their errands for a proficient and stringent improvement venture, various
moves must be made, incorporating an iterative advancement in little augmentations, early bug-location and
bug-settling, stable adaptation and setup administration, a strong engineering which grasps computerized tests at
any level of the product engineering, and the greater part of each of the, an altogether composed test foundation.

file:///C:/Users/IK/Desktop/SE/201701249.html 9/9

You might also like