This week I spent a good portion of my time ordering and receiving the final parts for the physical construction of our robot. Most of this was pretty straightforward, except we were having trouble finding a vendor with the iRobot Create 2 in stock. We eventually found a vendor in Missouri, and ordered the robot base from them. The majority of the parts arrived Friday, so I spent most of the day testing for part-to-part compatibility. The motors, wheels, and mounts all fit together, and the battery we ordered was able to power the motors. Although we did look over our list many times, we realized that we’ll need some additional parts. The LiPo battery we ordered did not include a charger, so we’ll need to put in another order for that. Additionally, the mounting hardware that connects the wheels to the motors uses very small hex screws, so we’ll need to find a properly sized allen wrench.
Before the parts arrived, I spent some time with the rest of the group working on the computer vision algorithm. I built all the necessary drivers and development APIs on my linux machine (since we plan on using linux onboard the robot) and experimented with CV algorithms for detecting tennis balls. After some fine-tuning, we were able to use our Intel RealSense camera to track multiple tennis balls. Ishaan’s update should have more details regarding the CV algorithm used to do this.
Now that most of the parts have arrived, I can begin to work on prototyping the mechanism for collecting tennis balls. I’d first like to get an idea for the motor speed required to grip and launch tennis balls. Then, I’d like to assemble a prototype structure to hold the motors in place and test the launching mechanism. After this, I’ll move on to prototyping the robot’s arms and ramp.