Recommendations for a robot for scanning in a 2D plane

Question

I'm new to robots and struggling to find information (buy) and sources on what we're looking for (or what we are trying to accomplish) for an idea (proof of concept) we want to test. We really don't want to build this ourselves, but we aren't willing to spend a lot since it's a proof of concept.

What we were looking to accomplish is to have a robot hold a scanning probe in a specific orientation as it scans through a 2D plane in space (no touch). Low noise and vibration are solid requirements. Speed is less important. We would like something that has some reach, maybe 4ftx4ft but its a POC...

I think there are a couple of designs that could accomplish this. For example, two linear systems for a simple xy positioning. More advanced would be 2 axis arm using servos or a multi-axis arm.

Ideally, some simple software to control it (motion planning where the operator can specify the size to measure and the number of scans in each axis - e.g. 30" by 30" with 30 vertical and 10 horizontal passes.

Everything I've seen is production ready and expensive so we're wondering if anyone is aware of a simple cost-effective system that might allow trying a few things?

Thanks in advance! PV

Answer 1

TL;DR: Robots are expensive.

The bigger the robot, the more expensive to build and maintain. The 4x4ft build volume is large for a static robot. Bigger robots endure higher stresses and require better building materials. Better building materials require better tools. If I had to guess, you probably saw a commercial robot that fit your needs north of ~\$30K to ~\$40K for an industrial robot with a 4x4 foot reach. You aren't going to get around that. Your best bet is to build a simple Scara style robot using parts from your local hardware store and motors from 3d printers (nema 23 steppers might have enough torque for your requirements). You might be able to make something for around ~\$1-2K if you do your research and shop for the best bang for your buck. Then, there is the lidar sensor itself. A good, basic lidar with good accuracy and repeatability and longevity will cost you at a minimum around ~\$1-2K, so you are looking at a budget of around ~\$5K just to get a hacked together prototype, assuming there is a minimal cost surprise. This also assumes you already own the tools and have the time to embark on such a project (not to mention consumables involved in construction). If you are new to robotics, your best bet is to find somebody near you that has more knowledge in the field to show you what to do. I am purposefully glossing over some of the ways you can minimize cost with component swaps, but then you risk your robot failing because it was built poorly from the start. All your design decisions have tradeoffs.

There is a glimmer of hope though. Some companies donate parts to educational institutions for student projects. I suggest you take advantage of this.