Lab 6: Seeing is Believing

Goal: Use computer vision to give your Pupper a sense of sight. At the end of this lab, Pupper should be able to track a person using its camera.

Step 1. Install X11 SSH into the Raspberry Pi

Step 2. Connect the Raspberry Pi to Stanford Wifi

Step 3. Install DepthAI on the Raspberry Pi

Use any of the Tiny YOLO models.

Step 4.

pseudocode for script 1: #stand in place pupper

find the center of the person boundary box using xmin + x max/2, ymin + ymax/2 find the distance of that center from (0.5, 0.5) using the distance equation assign the velocity to be proportional that distance * some constant that has to be tuned send the velocity to the other script

if no person is found, make the pupper return to center

script 2: receives the velocity and calls the pupper api update function for yaw control