This lab works on phones, but writing code is much easier on a laptop.
Perception · Modern Robotics Perception and Capstone · P9-L1 · Advanced
The new detector is more accurate on the benchmark and 50 ms slower. In the first rain test the robot drives into the pallet it would have seen with the old one.
You will turn latency into distance, solve for the speed the pipeline allows, and measure how each kind of fault eats into the detector’s recall.
You have passed every stage. Now connect it back to the theory and move on:
Plain Python with math (and slamkit for small matrix helpers: matmul, transpose, solve, inverse, identity). Images, point clouds and detections arrive as lists inside the dictionary D; an image is a list of rows, each row a list of pixels. There is no NumPy or OpenCV: you write the algorithm yourself, which is the point of the lab. Print with print().
Stage 1 explores latency, range and speed. Stage 2 computes the latency budget and the maximum safe speed. Stage 3 summarises a fault-injection experiment. Timings are stated numbers and the detections come from a simulator with a stated error model, not a trained network. Your code is graded on hidden pipelines and experiments.
Educational simplification: a sequential pipeline with fixed stage times, constant deceleration, reliable ranges that halve in rain, and three injected faults applied to simulated detections. Real systems have variable latency, shared compute, and many more failure modes. All data in this lab are simulated.
Get notified when new robotics labs are released. One email per new lab, nothing else.