Skip to content

Latest commit

 

History

History
148 lines (115 loc) · 4.7 KB

README.md

File metadata and controls

148 lines (115 loc) · 4.7 KB
Raw

_____ __ __ _________ __________ . _____ __
/ _ / \ / / / _ \ _ | ____ / \ _____ | | __ ___________ / /\ \ // /_ \ | // _ | __ \ / _ \ / \ / \_ \ | |/ // __ _ __
/ | \ / / \ | | ( <> ) _\ ( <> ) Y / __ | <\ /| | / _|__ /_/\ / /______ / || /_/|___ /_/_|__ (____ /|_ \_ >__|
/ / / / / / / / /

The IAM user needs read and write permissions to several services, including provisioning resources with AWS CloudFormation, AWS RoboMaker, Amazon S3, AWS Cloud9, Amazon CloudWatch, Amazon VPC, AWS Lambda and AWS Identity and Access Management. Please make sure you have these permissions in your account before proceeding further.

This description is valid for AWS Robomaker Cloud9 IDE and it is not tested locally(could work with minor changes)

Description

This repository is created to demonstarte CR2 robot on AWS Robomaker.

Initial ROS setup

  1. Download this repository
  2. Go to downloaded folder and run ./setup/ros_setup.bash It downloads all needed ROS dependencies and builds all ROS packages

Package description

cr2_control

Package for launching cr2 robot

robot_monitoring

Package for publishing different metrics(e.g. distance to the goal) to ros topics.

Run CR2 Single robot

To run single Catering Robot you need to run following command:

./setup/aws_setup.bash start_single_robot.json

During the first run it creates AWS infrustructure, copies bundle to the s3 bucket and runs simulation application.

Run CR2 Fleet

To run fleet on AWS Robomaker

./setup/aws_setup.bash start_four_robots.json

If AWS stack was not created before, script will create it.

Typical config to launch fleet

{
    "robots": [
        {
        "name": "robot1",
        "environmentVariables": {
            "START_X": "2",
            "START_Y": "2",
            "START_Z": "0.0",
            "START_YAW": "0",
            "USE_CUSTOM_MOVE_OBJECT_GAZEBO_PLUGIN":"true",
            "MAPPING":"false",
            "ROBOT_MONITORING":"true",
            "ROBOT_TESTING":"false"
        },
        "packageName": "cr2_control",
        "launchFile": "cr2_fleet.launch"
        }
      
    ],
    "server": {
        "name": "SERVER",
        "environmentVariables": {
            "START_X": "0",
            "START_Y": "1",
            "START_Z": "0.0",
            "START_YAW": "0",
            "USE_CUSTOM_MOVE_OBJECT_GAZEBO_PLUGIN":"true",
            "MAPPING":"false",
            "ROBOT_MONITORING":"true",
            "ROBOT_TESTING":"false"
        },
        "packageName": "cr2_control",
        "launchFile": "cr2_fleet.launch"
      }

  }

There are several environment variables to configure simulation:

  1. MAPPING - enable mapping using gmapping and explore_lite, default - false
  2. ROBOT_MONITORING - enable robot monitoring, default - false
  3. ROBOT_TESTING - run tests (on current version simple navifation test), default - false

CICD

This repository is ready for AWS CICD using AWS Pipelines

Steps to configure CICD:

  1. Create AWS stack
./setup/aws_setup_cicd.bash

This script will create two stacks: first contains only bucket for storing bundles and AWS lamdas sources. The second stack creates all needed infrustructure for running simulation jobs. 2. Update build file buildspec.yml with your S3Bucket name received in first step, so build machine will know where bundle should be located

    S3_BUCKET: <YOUR-BUCKET>
  1. All further instructions how to configure AWS CICD Pipeline you can find here https://github.com/aws-samples/aws-robomaker-simulation-launcher

Note! Build role must have next access to the S3Bucket or use one created by cloudformation(name contains CodeBuildRole)

        {
            "Effect": "Allow",
            "Resource": [
                "arn:aws:s3:::<YOUR-BUCKET>/*"
            ],
            "Action": [
                "s3:PutObject",
                "s3:GetObject",
                "s3:GetObjectVersion",
                "s3:GetBucketAcl",
                "s3:GetBucketLocation"
            ]
        }

You can configure CICD scenarios in file "scenarios.json"

Cleanup

To delete the sample application use the AWS CLI.

For Fleet application

aws cloudformation delete-stack --stack-name r2-robomaker-fleet

For CICD

aws cloudformation delete-stack --stack-name r2-robomaker-fleet-cicd
aws cloudformation delete-stack --stack-name r2-robomaker-fleet-cicd-bucket