THE 3 rd INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS dedicated to the 99
th
anniversary of the National Leader of Azerbaijan Heydar Aliyev
197
SELF-DRIVING CARS BASED ON CLOUD COMPUTING Fuad Mammadzade, Kanan Bagaliyev, Turan Eminli Baku Higher Oil School Baku, Azerbaijan Academic supervisor: Ph.D, Assoc. Prof. Kamala Pashayeva Keywords : Real-time data transfer, Self-driving cars, Cloud Robotics
In modern multi-robotic systems research, which is a relatively new field,
their implementation on cloud systems and utilizing the out-of-the-box
features of cloud resources are widely investigated for their importance on
task managing, data logging, and getting precise predictions for autonomous
driving. The major objective of utilizing the system based on cloud computing
is to alleviate the complexity of distributed computing and storage and
heterogeneous computing. Each self-driving car generates enormous data
from different sensing elements such as GPS for HD map generation, LiDAR
for 3D mapping the environment and understanding obstacles, Camera for
all the computer vision operations from sign detection to the road line
determination and MPU accelerometer or Inertia Measuring Unit for detecting
the behavior of the robotic system and car.
[1] According to the project described in this paper, the different robotic
agents or cars will detect different problems or advantages along the route,
which is determined beforehand, with their sensors and they will transmit this
collected data to the cloud platform for processing on a real-time basis. The
collected data will assist other robotic systems to behave smart and more
robustness could be achieved in this approach. The grid that has a
dependency on remote servers, will create huge information and availability
for predicting actions and defining the correct path. With the infrastructure of
hardware and software introduced in this project, we will be able to utilize the
real-time data on the autonomous cars with remote control and the
instructions will be generated according to the data which is in the cloud
platform and gained from other agents’ sensor data contribution.
If a project involves some kind of vehicle, it is necessary to measure
physical quantities like acceleration, speed, traveled distance and rotation
angle. We would need several measurement devices in order to measure
these quantities. First, we need an accelerometer to measure acceleration in
3 different axes. Accelerometers have many uses in industry and science.
Highly sensitive accelerometers are used in inertial navigation systems for
aircraft and missiles. Vibration in rotating machines is monitored by
accelerometers. They are used in tablet computers and digital cameras so
that images on screens are always displayed upright. In unmanned aerial
vehicles, accelerometers help to stabilize flight.