AWS ‘Smart Farm’ : ML, Computer Vision and Data Streams

AWS whitepapers on making “Smart Farm” by using multiple AWS services with ML, Computer Vision and Data Streams.

Refer to the below architectural diagram: (AWS Services used in this architecture are explained below)

Enabling sensor, computer vision, and edge inference in agriculture.

Source: AWS Whitepaper

Ingredients for this Farm Fresh Pizza, NO :-P I mean ‘Services’ used in this AWS solution.

1. AWS Lambda: It enables you to run code without running the servers. You can setup trigger from other AWS services or call it directly via apps.
2. AWS Simple Storage Service Aka. ‘S3’: An object storage service that lets you segregate your data into different buckets.
3. AWS SageMaker: An service that provides you ability to build, train and deploy you ML models.
4. AWS IoT Greengrass: This service seamlessly extends AWS to edge devices so that they can act locally on the data they generate, while still using the cloud for management, analytics, and durable storage.
5. AWS Direct Connect: A solution that makes it easy for you to establish a dedicated network connection from your premises to AWS that helps you provide better bandwidth throughput with consistent network at lower prices.
6. AWS PrivateLink: AWS Service that provides private connectivity between VPCs, AWS services, and on-premises applications, securely on Amazon network.
7. Amazon Kinesis Data Streams(KDS): It is highly scalable and durable real-time data streaming service that can enable up to capture Gbps of data per second from end number of sources.
8. Amazon Kinesis Video Streams: This service enables you to playback video for live and on-demand viewing, and build applications that run video analytics on it.
9. Amazon Kinesis Data Firehose: It loads up data streams directly into AWS products for processing with auto scaling. It also allows streaming of data to S3, Elasticsearch Service, or Redshift as per which the data can be analyzed via other services or applications. Read More
10. AWS IoT Core: A managed cloud service that enables connected devices easily and securely interact with other devices and cloud applications.
11. AWS IoT Device Management: It lets you track, monitor, and manage connected IoT devices.
12. AWS IoT Device Defender: A service that helps you secure your fleet of IoT devices.
13. Amazon Simple Notification Service (SNS): It is a messaging service that enables messaging between different services via SMS and Email.
14. Amazon API Gateway: It is a fully managed service that makes it easy for developers to create, publish, maintain, monitor, and secure APIs at any scale. APIs act as the “front door” for applications to access data, business logic, or functionality from your backend services. Source
15. Amazon Elasticsearch Service: Service that lets you run Searches for getting the in-depth information about your infrastructure.
16. Amazon QuickSight: A cloud powered Business intelligence service that lets you easily create and publish interactive dashboards that can then be accessed from any device or location with internet.
17. Amazon FreeRTOS: It is an open source, real-time operating system for microcontrollers. Example: Smart Watch or Smart Car.
18. Sensors: Camera and other required sensors as per your use case.

Architectural Explanation:

1. Using 3rd party sensors or drones not using FreeRTOS to send data via AWS Lambda for protocol conversion.
2. Sensors or cameras running FreeRTOS send data to AWS IoT Greengrass providing protection from intermittent connectivity.
3. AWS IoT Greengrass streams enables ingestion from edge devices to Amazon Kinesis Data Streams.
4. Use real-time video via Amazon Kinesis Video Streams for streaming and replay of video content.
5. Derive real-time insights with Amazon Kinesis Data Analytics and notify users via Amazon SNS.
6. Enable analytics with Amazon Elasticsearch Service and use Amazon S3 for a data lake strategy.
7. Transfer owned data like planting records or farm finances securely into your data lake with AWS Direct Connect.
8. Consume data from a sensor ecosystem hosted on AWS securely with AWS PrivateLink.
9. Empower users with insights delivered via Amazon API Gateway or visualizations with Amazon QuickSight.
10. Build and deploy Machine Learning (ML) models for edge inference with Amazon SageMaker.

Done!

Source~

I am an avid reader and love to spend my time reading about different technologies. If you want to connect for any of your technology discussions or stories, feel free to reach out as I would love to know more about your technology stacks. :).

~AshishSecDev