Build smart robots and operate them easily with AWS Robotics
summary
Shows concrete stacks and outcomes: uses AWS IoT Greengrass with a Stream Manager component to ingest sensor data into Kinesis and S3, deploys SageMaker-trained models to robots for edge inference, and establishes low-latency teleoperation via Amazon Kinesis Video Streams with WebRTC. Demonstrated deployments include a cloud-trained object detector running on Boston Dynamics Spot, Agrointelli’s entire fleet managed remotely, Seafloor Systems saving hours per USV and apps deploying in minutes, and iRobot running hundreds of simulation tests for five million units using RoboMaker.
AI-written summary · about
chapters (15)
- 0:00Intro
- 0:21Robotics is key to digital transformation
- 0:55Robotics industry trends
- 3:53AWS loT services: Foundation for cloud robotics
- 6:05Data ingestion, storage, and analytics
- 9:49Video streaming for teleoperation & supervision
- 11:51Machine learning inference at the edge The challenge
- 12:49Machine learning at the edge solution
- 14:51Software provisioning and OTA updates The challenge
- 15:33AWS IOT Greengrass 2.0 for robot software provisioning and OTA
- 16:50Deploying and managing software on robots
- 19:16Fleet monitoring and dashboards
- 21:38Robot software development cycle
- 22:45Simulation with an API using AWS RoboMaker
- 26:52AWS Training and Certification
related talks
30mBuilding, testing and operating robots with AWS
43mPatterns for cloud-enabled robots
18mHow Amazon builds and deploys autonomous robots
45mBuilding Interactive, AI-powered IoT Applications
26mInnovate with AWS: Enabling industrial ops at Amazon fulfillment centers
17mTransforming Robotics Service with Agentic AI on
36mRobots and assistive technologies
48mBuilding Robotics Applications at Scale using Open Source from Zero to Hero