Low-cost 4-DOF robotic arm using ROS: collision-free path planning
thesisposted on 28.03.2022, 13:00 authored by Jacob Blanck
Robotic arm control and path planning is a highly important aspect of research due to the rise in robotic arm applications and advancements in their capacity to work accurately and safely. ROS software has been used to achieve this collision-free path planning and robotic arm control in previous research, however these robotic arms are often expensive and supply ready made SDK packages that function within ROS. In this project, ROS software is used to simulate the "AX-12A Smart Robotic Arm" from CrustCrawler. This low-cost robotic arm has limited open-source ROS software associated with it, in addition to no accessible simulations of the robotic arm. The aim of this project is to form a collision-free static path planning model within ROS, similar to packages available for more expensive robotic arms.The ROS software generates a random goal state and plans a collision-free path for the robotic arm to follow, achieving the desired state. This is achieved through a MoveIt! configuration generated from SRDF and URDF files created for the robotic arm. The results show that the URDF functions correctly and the MoveIt! configuration path plans successfully. The simulation is able to identify configurations that include self-collision and will not path plan to this collision. This research is significant for the use of low-cost robotic arms within both research and and industrial industries while accurately executing path planning to a goal state and operating without collision.