Description

ABSTRACT

Machine learning techniques have been widely applied, achieving state-of-the-art results in various fields of study. This survey focuses on machine learning solutions that target learning control policies for robotics applications. Autonomous robotic systems may be seen as a conjunction of different subsystems as sensing, control, and decision making. Traditionally, these subsystems have been modelled from first principles, and solutions were implemented based on these models. There are several reasons for using a model-based approach and among them is risk reduction. Designers need to guarantee some safety boundaries for robots (or any other physical dynamic system). If we know the model of a system, we can design sensing, control, and decisionmaking subsystems to perform the task or mission we want with reasonable guarantees of safety.The performance of the robot is validated with the simulation results and two physical experiments. The results show that the robot could navigate autonomously in unknown environments. 

EXISTING SYSTEM

• In the existing system, there is only the sensors will be used to control the robot.
• Using the Conditions based Data driven method.
• Only detects the obstacles range using sensors.

PROPOSED METHOD

• Open CV based image processing algorithms are used to predict the obstacles.
• Machine learning algorithm based Data driven method.
• User-friendly Programming language 

BLOCK DIAGRAM

HARDWARE REQUIREMENTS

• Raspberry Pi 3
• Camera module
• GPS
• Ultrasonic sensor 

SOFTWARE REQUIREMENTS

• Operating System – Raspbian jessie
• Programming language – Python
• Open CV with Imagga
• Machine learning