2D Simultaneous Localization and Mapping (SLAM) based Autonoumous Rover with Environment Surveillance.
Advisor:
Mr. Arsalan A Rahim
Abstract:
The autonomous navigation of robots in unknown and uncontrolled environment is a huge
challenge because it requires a set of subsystems to work together. Simultaneous Localization and
Mapping (SLAM) based robot addresses the problem of creating a map of an unknown
environment having static or dynamic landmarks and simultaneously locating its own position in
the map. Localization and mapping are interdependent so doing both simultaneously is a difficult
task. Laser Sensor is being used to measure the distance of the objects in the environment. The
Gyroscope and speed-sensor are used for odometry reading to estimate the approximate position
of the rover in the environment. Occupancy Grid Mapping Technique is used to map the
surrounding environment in which map is divided into grids having different weights based on
occupied and free space. Particle Filter based Localization is used to localize and find the actual
most certain position of the rover. In Particle Filter, many instances of the rover are created and
their correlation with the map is calculated to find the best position. A solution to the SLAM
problem would help a robot in forming a complete map of the environment without any human
assistance. Once it generates the map, it can choose how to traverse the surroundings by
interpreting the map with the help of particle filter localization, or a location can be chosen at the
users end for where the robot should go. The robot also has environment monitoring capabilities
using Temperature sensor and Luminosity sensor. These sensors provide information about the
physical conditions of the environment such as the temperature of surrounding objects, ambient
temperature and intensity of light. The design of the structure itself is to allow it greater
maneuverability in rough or uneven terrains. This is made possible by using Rocker-Bogie
mechanism. A SLAM based robot has applications in many fields like self-driving cars, package
delivery, cleaning, agriculture, surveillance, search and rescue, transportation. Note that all these
applications occur in uncontrolled and unknown environments.