examples/robotics.md

GitHub

Robotics

Robot fleet orchestration with RoboMaker simulation, Greengrass edge runtime, and cloud coordination.

Key Elements

Component Stencil
RoboMaker mxgraph.aws4.robomaker
RoboRunner mxgraph.aws4.iot_roborunner
Greengrass mxgraph.aws4.greengrass
ML Component mxgraph.aws4.iot_greengrass_component_machine_learning
IoT Core mxgraph.aws4.iot_core
Camera mxgraph.aws4.camera
S3 mxgraph.aws4.s3

Example

Warehouse robot fleet with simulation, edge ML, and centralized coordination:

@startuml
left to right direction

rectangle "Warehouse Floor" {
  rectangle "Robot A" {
    mxgraph.aws4.camera "Vision\nCamera" as cam1
    mxgraph.aws4.greengrass "Greengrass\nRuntime" as gg1
    mxgraph.aws4.iot_greengrass_component_machine_learning "Navigation\nML" as ml1
  }
  rectangle "Robot B" {
    mxgraph.aws4.camera "Vision\nCamera" as cam2
    mxgraph.aws4.greengrass "Greengrass\nRuntime" as gg2
    mxgraph.aws4.iot_greengrass_component_machine_learning "Navigation\nML" as ml2
  }
}

rectangle "Cloud Orchestration" {
  mxgraph.aws4.iot_core "IoT Core" as core
  mxgraph.aws4.iot_roborunner "RoboRunner\nOrchestrator" as runner
  mxgraph.aws4.robomaker "RoboMaker\nSimulation" as sim
  mxgraph.aws4.s3 "ML Models\n(S3)" as s3
}

cam1 --> gg1
gg1 --> ml1
cam2 --> gg2
gg2 --> ml2
gg1 --> core : status
gg2 --> core : status
core --> runner
runner --> sim : "test scenarios"
s3 ..> gg1 : "model deploy"
s3 ..> gg2 : "model deploy"
@enduml

Pattern Notes

  1. Edge ML per robot: each robot runs Greengrass with local navigation ML — no cloud latency for real-time decisions
  2. RoboRunner: centralized task orchestration across heterogeneous robot fleets, manages task assignment and routing
  3. RoboMaker simulation: test new navigation algorithms in virtual warehouse before deploying to physical robots
  4. Model deployment: trained models stored in S3, pushed to robot fleet via Greengrass deployment
Icon Stencil Use When
Kinesis Video mxgraph.aws4.kinesis_video_streams Streaming robot camera feeds
Rekognition mxgraph.aws4.rekognition Visual object/defect detection
SageMaker mxgraph.aws4.sagemaker Training navigation models
IoT FleetWise mxgraph.aws4.iot_fleetwise Multi-robot fleet telemetry
Device Mgmt mxgraph.aws4.iot_device_management Robot provisioning & grouping
Industrial PC mxgraph.aws4.iot_thing_industrial_pc On-robot compute hardware