Skip to content

UCR-Robotics/ICRA-RM-Sim2Real

 
 

Repository files navigation

ICRA-RM-Sim2Real

RoboMasterEP

The RoboMaster EP chassis

RMUS-playground

The RMUS playground

This repository is dedicated to provide the essential information for developpers to participate the RoboMaster University Sim2Real Challenge.

The github repo and gitee mirror are provided for various users:

ICRA-RM-Sim2Real Tutorials

arch of the RMUS EP SIM2REAL system

The RMUS SIM2REAL system consists of the Server/Client two parts seperately, one is the server docker image to deploy the simulation environment and the other is the client docker image to deploy the user algorithm for the EP robot to connect the environment both for simulation and reality.

Non-Beginners: If you're already familiar enough with Linux+ROS+Habitat or earlier versions and only want to explore the new ICRA-RM-Sim2Real playground, you can go through the details of installation guide here. However, going through all basic Entry Level tutorials is still recommended for all users to get exposed to new features.

If you are new to Linux: You may find it helpful to first do a quick tutorial on common command line tools for linux. A good one is here.

Hardware requirement: NVIDA GPU 1080+ with driver 470+ in need to support the Sim2Real environment.

Ubuntu 20.04 or later in need to support more convenient experience.

Entry Level

Intermediate Level

Challenge overiew

FPV-RMUS

The RoboMaster University Sim2Real Challenge (Sim2Real) at its core allows participants to win points by rearranging mineral using fully automated RoboMaster EPs that have been modified officially.

The match lasts for five minutes with a sim2real-based format, where robots rearrange minerals based on the information shown on the exchange tags to earn points. Participants will be ranked according to their total points won.

pick-ore

place-ore

The objective of the challenge is to assess how well a program completed on a simulation platform can be operated in real application environments.

Teams are required to develop and debug their algorithms in a simulator and submit their codes by the specified deadline. The official staff will deploy corresponding codes in physical robots of the same models to earrange minerals.

Compared to other RoboMaster University events, this challenge does not require teams to build physical robots, allowing participants to focus entirely on algorithm design.

Other resource

Related docker repository

Reference

  1. ROS-X-Habitat: Bridging the ROS Ecosystem with Embodied AI
  2. https://nvidia.github.io/nvidia-docker/
  3. How to Train Your [Dragon] Embodied Agent (with AI Habitat)
  4. https://github.com/facebookresearch/habitat-lab
  5. https://github.com/ros/dynamic_reconfigure

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Shell 100.0%