This week we were very productive and have been preparing for the interim demos. We worked more on robot construction and have finished the frame and ramp with the intake attached to the front. We still haven’t found the perfect motor yet for the intake, but we are very close and know what we need to order now. Below is a picture of the robot now. There is also a top piece of lexan attached that will hold the camera, Jetson, Arduino and motor driver.
On the software side, we integrated inference, distance, and angle calculation to center the robot to the target. We hope to test angle calculation with the physical robot tonight. Inference is still running very slowly (20s per frame), mainly due to cuda not being available, so we hope the software pipeline time will decrease dramatically after resolving cuda issues. We believe we are on track for hardware and slightly behind schedule for software.
0 Comments