Clarification on Docker Image Size & Multi-Environment Setup (Local Pass, Submission Fails)

Competitions AI for Industry Challenge
1

Hi,

I’m encountering an issue where my solution works correctly in local testing, but fails during submission/evaluation, and I’d like to clarify a couple of points.

  1. Docker image size
    My current submission image is quite large (~56GB).

  • Are there any explicit or implicit limits on image size for submission?

  • Could a large image cause failures during upload, deployment, or evaluation?

  1. Multiple environments / processes
    In my setup:

  • The ROS 2 lifecycle node (aic_model) runs in one environment

  • The model inference runs in a separate environment (launched via subprocess)

This setup works reliably in local tests, and all lifecycle and action behaviors appear correct.

I would like to confirm:

  • Is this multi-environment / multi-process setup allowed?

  • Are there any restrictions on spawning subprocesses or using multiple Conda/Micromamba environments inside the container?

Any guidance would be greatly appreciated, as I’m trying to understand why a locally working solution fails after submission.

Thanks!