How to Build a Robot Application Webinar

How to Build a Robot Application is a free Design News webinar that will air on Thursday, August 22 at 2:00 pm Eastern.

The webinar will explore the steps involved in designing an effective robot application in a production environment. Beginning with the essential concept – whether it’s a pick-and-place application or a depalletizing operation – you will have to make decisions on the type of robot you need and how to configure the robot to complete its tasks. The webinar will consider a wide range of robots, end effectors, and the different methods for coordinating robot motion control.

Presenter John Blyler is the founding systems engineer for JB Systems. He has designed, built, tested, and operated robotic systems for the military, commercial, and manufacturing industries. He has hands-on experience as an embedded systems engineer and project manager.

We caught up with Blyler to get details on what he plans to cover during the webinar. He explained what questions are involved in identifying the type of robot you need. He also explained the process involved in deploying a robot application.

How do you decide what type of robot to add to accomplish your production goals?

John Blyler: The first step is to understand what the robot needs to do. Will the robot scurry around the floor? Is it fixed on an assembly line for cars? Will it help in building something like a PCB? You need to know your own strengths. Do you buy a system that is already set up, so you don’t have to do hardware or software? Do you have expertise in end effectors? Are you adding your first robot or are you adding to an existing system?

Once you know what you want to do, look at your strengths and weaknesses. Then you design the application based on your needs and capabilities. Do you need a special arm? Do you need a cobot that works with humans? Based on this, you make a list of requirements for hardware and software.

What is involved in implementing the robot application?

Blyler: Moving on to implementation, do you need embedded cameras? What sensors will be required? What control system will you use These are basic engineering questions that help identify your problem.

Then you figure out how much memory you will need. Will your control be in the cloud? Do you do it all in software? You will probably have to add hardware. Do you need sensitivity or position force. After that, you move to test scenarios. Maybe you do some simulation first. Simulation can lead to a digital twin that is part design and part test.

AI is starting to get used to train robots to do things in this space. AI can be paired with innovative hardware design or operational approaches, such as robots that run along the ceiling in elderly nursing facilities. These scenarios require new behaviors by the robotic system, and that’s where AI comes in. (See, “A Glimpse at Toyota's Robot-Amplified Future”)

What is involved in the overall robotics automation system?

Blyler: An automated robotic system – from automotive lines to PCB manufacturers - involves a combination of hardware, software, and processes to perform tasks with minimal human intervention. Many modern lines add cellular manufacturing sublines to product products with variations, but an overall systematic process must be intact to confirm the inputs and outputs for each cell. A big-picture view will also help to reduce power and overall costs. The cells need to be designed to work independently, but you need an overreaching system for total quality. In a factory, you’re trying to optimize speed and efficiency. You need to be aware of these and other requirements.