Flaky test: test_go_to_the_bed (test_dimsim_spatial_memory) times out intermittently

Flaky Test: test_go_to_the_bed

File: dimos/e2e_tests/test_dimsim_spatial_memory.py
CI Run: https://github.com/dimensionalOS/dimos/actions/runs/27148687897/job/80133497365

Behavior

The test fails on first run with a 180s timeout waiting for the robot to reach the target position, then passes on retry (same run). Classic flaky test.

Error

FAILED dimos/e2e_tests/test_dimsim_spatial_memory.py::test_go_to_the_bed - TimeoutError: Failed to get to position x=-3.567, y=-1.332

dimos/e2e_tests/lcm_spy.py:105: TimeoutError

    human_input("go to the bed")

>   lcm_spy.wait_until_odom_position(-3.567, -1.332, threshold=2, timeout=180)

E   TimeoutError: Failed to get to position x=-3.567, y=-1.332

Retry (same run)

run-last-failure: rerun previous 1 failure (skipped 216 files)
dimos/e2e_tests/test_dimsim_spatial_memory.py::test_go_to_the_bed PASSED [100%]
415.14s call     dimos/e2e_tests/test_dimsim_spatial_memory.py::test_go_to_the_bed

Analysis

• The test sends "go to the bed" and waits for odom to reach (-3.567, -1.332) within threshold=2, timeout=180s
• On retry it took 415s to pass — well over the 180s timeout on first attempt
• The 180s timeout is likely too tight for this agentic navigation test (LLM reasoning + spatial memory lookup + path planning + execution)
• Non-determinism in LLM responses and exploration timing makes this inherently variable

Suggested Fix

• Increase timeout from 180s to ~600s (matching pytest-timeout), or
• Add a retry decorator for known-flaky agentic tests, or
• Split the test: verify the navigation command was issued (deterministic) separately from verifying arrival (non-deterministic)

Synced from DIM-996 by summer