Path Tracing in holonomic drive system with Reduced Overshoot using rotary encoders

EasyChair Preprint 2067

Abstract

This paper focuses on path tracing of a holonomic three wheel Omni drive system. Any path (curved, linear) is first sampled, where the coordinates of the path are stored in the local memory, at particular sampling rate which are then retrieved dynamically during the runtime to move the robot on the similar path. The task of moving the drive system from one coordinate to another is done by Deduced reckoning algorithm. This paper also proposes algorithm for minimizing overshoot (deviation from actual path) while moving in a curved trajectory. The proposed algorithm helps in keeping track of the robot very efficiently. The hardware employed to test the proposed algorithm used consists of a three wheeled Omni base, three high torque DC motors, three 6” Omni wheels, highly efficient DC motor drivers, high precision rotary encoders and a microcontroller powered by ARM® Cortex®-M3 processor. The robot controlled by the proposed algorithm was tested on wooden field while plying on curved paths.

Keyphrases: Deduced Reckoning, Holonomic Drive, IMU, Rotary Encoders, omni direction

@booklet{EasyChair:2067,
  author    = {Divyanshu Tak and Ayush Jain and Paras Savnani and Akash Mecwan},
  title     = {Path Tracing in holonomic drive system with Reduced Overshoot using rotary encoders},
  howpublished = {EasyChair Preprint 2067},
  year      = {EasyChair, 2019}}