RL Robotics LabSearch ↗
Lesson / authored

Week 22 · Improve: A baseline should be reproducible

Session 4 of 4 · Integrate the system · Phase 6

Plan about 15 minutes for explanation, 30 minutes for practical work and 10–15 minutes for documentation. A longer build may continue into the next session: stop safely, commit the current state and record the next check. Desktop simulations count as software evidence; label them clearly and record physical validation separately.

Engineering challenge

Can every module cooperate without weakening the stop and measurement contracts? This session focuses on a baseline should be reproducible.

Before you start

The previous week’s recorded baseline and Week 21, Improve. For later sessions this week, retain the preceding session’s files and predictions.

Equipment: Desktop Python, editor, paper and ruler; for physical work, the configured 3pi+ 2040, clear floor mat and hardware checklist. Week 3 additionally uses the separate low-voltage LED circuit described in its procedure.

For any motion, verify the stop button, short time limit and clear floor area first. Keep the wheels raised for a new device program until its commands and stop behaviour are checked. A hazard or uncertain input is a reason to stop and document, not to force the trial to finish.

Theory and mathematics

A baseline should be reproducible

A working integration build consists of source, configuration, map, dependency/firmware versions, start procedure and retained evidence. Record the Git revision before a run. If the robot contains uncommitted uploaded changes, mark the run as exploratory until you recover the exact source. Replaying a log should reproduce decisions where inputs and timing are fixed. A simulator checks logic under its assumptions; it cannot establish motor torque, electrical reliability or floor traction. Keep those validation claims distinct in the report.

Worked example — illustrative values

Consider a sample with goal_distance = 0.02 m and bumper_pressed = True. Even if 0.02 m is inside the arrival tolerance, hazard priority selects STOP_HAZARD. With a 60 s mission budget, an elapsed time of 60 s selects STOP_TIMEOUT. A lower-priority target must not overwrite either decision.

Write the calculation in your notebook before running code. State which values you measured, which you assumed and which the program calculates. A correct numerical calculation cannot rescue an incorrect physical assumption.

Run and explain the model

The following is desktop Python, not a ready-to-run motor program. Download this week’s example, save it in your student repository and run python3 code/w22.py from the repository root. The same small model is reused across the week so you can learn it, build with it, test it and revise it.

# Desktop Python teaching example. Numerical inputs are illustrative.
def arbitrate(*, fresh, hazard, elapsed_s, arrived):
    if hazard:
        return "STOP_HAZARD"
    if not fresh:
        return "STOP_INVALID"
    if elapsed_s >= 60:
        return "STOP_TIMEOUT"
    if arrived:
        return "STOP_ARRIVED"
    return "FOLLOW"

for case in [(True, True, 2, True), (False, False, 2, False),
             (True, False, 60, False), (True, False, 2, True)]:
    fresh, hazard, elapsed, arrived = case
    print(arbitrate(fresh=fresh, hazard=hazard, elapsed_s=elapsed, arrived=arrived))

Compare the unchanged baseline and your proposed change on the same inputs. Keep the original files so another reader can reproduce the comparison. If an exception appears, read its final line, identify the input or assumption that caused it and make the smallest explained correction. Do not delete validation merely to obtain output.

Understanding the model and its limits

The arbitration function gives hazard priority, followed by invalid data, timeout and arrival. Thus simultaneous arrival and contact reports STOP_HAZARD. The sample returns decision strings; the device adapter’s single motor owner converts any stop decision into motors.off and prevents stale targets from being reused. Test combinations, not only one condition at a time. A later navigation write can undo a stop unless motor ownership is explicit. Record the input sample and selected reason so replay can identify the exact decision boundary where the system diverged.

Practical instructions

  1. Fix the most consequential observed defect and rerun the entire small fault matrix.
  2. Freeze a baseline version with map/settings and a setup checklist.
  3. Repeat three short route runs with no changes between them.
  4. Tag the student repository baseline and prepare the independent Week 23 validation batch.

Experiment

After the fix, rerun eight cases and three baseline routes; preserve earlier failed-version evidence separately.

Before testing, record your prediction, changed factor, measured response, fixed conditions and stopping rule. Save every attempted run, including failures, with a condition and source version. If hardware is unavailable, use an explicitly labelled synthetic/replay dataset and list the physical question it cannot answer. Do not invent completed trials.

Deliverable

A frozen reproducible baseline, three retained route outcomes and a separate validation plan.

Save notebook/w22-s4.md, the relevant code revision, raw CSV or test-case records, and one labelled diagram/plot/table. Link the files relatively from your notebook. Use the entry template and report guide.

Completion criteria

A documented failed prediction can meet the learning criteria. A missing physical trial must remain marked untested; software success alone does not validate the robot.

Reading and video

Reflection and next step

Which assumption most affected your result? Point to one observation that supports your explanation and one alternative explanation the evidence has not ruled out. Write a specific next test with a changed factor and measurable outcome, then proceed through the week’s Learn → Build → Experiment → Improve cycle.

← Previous sessionCourse roadmapNext session →