Mohammad Zainullah Khan
Download Project (1.0 MB)
With advances in technology, robots have become an integral part of industries across the board. They are being employed in all sorts of applications from simple pick and place operations to the more complex operations involving AI with computer vision. In the manufacturing sector, robots are mostly used to perform tasks in isolation. There lies a big opportunity for efficiency improvement by having robots collaborate on tasks. This brings along with it challenges of obstacle avoidance, task allocation, and deadlocks. These challenges are easier to tackle for non-varying tasks such as a multi-robot system (MRS) used for spray painting a specific part. But when the same MRS has to be used for a number of tasks such as spray-painting a wide variety of parts, each with its own requirements, the challenges become harder to solve. The goal of this research is to advance task allocation and trajectory planning for multiple robot agents working collaboratively to perform manufacturing tasks. These industrial robots can vary from simple gantry robots to industrial robot arms mounted on mobile bases. Their applications will involve low-volume, high-mix manufacturing tasks such as spray painting, pressure washing, 3D printing, media blasting, and sanding. Apart from dealing with the generation of an offline collision-free path, manufacturing constraints must be considered as well. These involve achieving a constant speed of end-effector throughout a trajectory to avoid any undesirable effects. This research focuses on developing a technique for several robots with 3 or more revolute and/or prismatic joints with partially shared workspaces that enables them to allocate and perform manufacturing tasks in a time-effective and computationally efficient manner.
Andrew Murray, Dave Myszka
Primary Advisor's Department
Mechanical and Aerospace Engineering
Stander Symposium, School of Engineering
"Task Allocation and Dead-Lock-Free Trajectory Planning for Collaborative Multi-Robot System" (2023). Stander Symposium Projects. 2889.
Presentation: 9:00-10:15 a.m., Kennedy Union Ballroom