For the final exam of the course "Robotic Perception and Action" a project has been realized consisting of the kinematic calibration of a robot. This is a study on the odometric reconstruction of a moving robot starting from four datasets.
First task the cleaning of the cleaning of the data sets has been done by the duplicated information. This task was accomplished by executing a Python script that by providing data input files deletes duplicate records and rewrites a new file with the right information. Pandas package was used to generate dataframe.
Second task: In Matlab environment was performed the kinematic calibration using classical equations of odometry to determine the geometric parameters of the robot. Finally using the gentenetic algorithms was performed an optimization voted to determine the correct parameters of robot.
The report is avaible here with all details clearly exposed.
This project is licensed under the MIT License - see the LICENSE.md file for details