Problem

The task was to develop a robot capable of navigating through a predefined environment while avoiding collisions with fixed objects. The project required the integration of MATLAB and Arduino to develop algorithms that would enable the robot to perceive its surroundings and compute optimal navigation paths.

The challenges faced was creating an autonomous system that could process images in real-time and make decisions based on that data. it was essential that the robot could navigate through the environment without colliding with any obstacles, while also being able to adapt to changes in the environment.

Solution

To address this problem, a system was developed using MATLAB and Arduino that allowed the robot to perceive its surroundings and compute optimal navigation paths in real-time. MATLAB's image processing capabilities were used to detect obstacles in the environment, which were then communicated to Arduino for real-time control of the robot's movements.

Implementation

The implementation phase involved building a robot from the OSEPP Tank-01 mechanical kit. this included putting the chassis together and adding necessary components such as DC motors and motor drivers, a microcontroller and ultrasonic sensors in order to make the robot autonomous and susceptible to the programming algorithms.

In the project, MATLAB was employed in conjunction with an ultrasonic sensor to facilitate accurate and real-time distance measurements, enabling the robot to perceive its surroundings. The ultrasonic sensor emitted sound waves, which, upon hitting obstacles, bounced back to the sensor. MATLAB was programmed to process these echo signals, calculating the time taken for the sound waves to travel back and forth. Applying the speed of sound as a constant factor, MATLAB converted these time measurements into precise distance values. These distance values served as data for the robot's decision-making algorithms, allowing it to detect obstacles and determine their proximity in order to adjust its navigation path in real-time to avoid collisions effectively.

The Arduino code, in C++, implemented a decision-making algorithm that processed the distance data in real-time. When an obstacle was detected within a predefined range, the code triggered specific actions such as altering the robot's direction or stopping altogether to avoid a collision. The code was constantly tested and refined, considering various scenarios and obstacle configurations. Fine-tuning the code was essential to achieve a balance between swift responsiveness and stability.

Skills Developed

MATLAB
Arduino
C++
Mechatronics

Autonomous Robot to Navigate Complex Environments