Android Things

Android Things
Name	Android Things
Developer	Google
Released	2018
Discontinued	2021
Programming language	Java, C, C++
Operating system	Embedded Linux (modified)
License	Apache License 2.0

Android Things Android Things was a Google-developed embedded operating system intended to enable developers to build Internet of Things devices using familiar Android tools and services such as Google Play Services, TensorFlow Lite, and Firebase. The platform aimed to bridge Qualcomm-based system-on-module hardware, NXP Semiconductors chips, and partners like ASUS in order to simplify manufacturing, integrate with Google Assistant, and leverage cloud services from Google Cloud Platform. Initially announced at events like Google I/O and showcased alongside projects from Intel and MediaTek, the project later shifted focus toward commercial customers and was discontinued in favor of other embedded strategies.

Overview

Android Things provided a managed OS image built on a modified Linux kernel with an emphasis on long-term security updates, integration with Android Studio tooling, and support for common IoT protocols utilized by vendors such as Broadcom, Texas Instruments, and NXP Semiconductors. The platform targeted categories including smart speakers, digital signage, and industrial controllers produced by companies like OEMs such as Lenovo, Sony, and Motorola partners, and it promoted compatibility with cloud platforms like Google Cloud Platform and analytics services from Firebase Analytics. By reusing components from Android and services from Google Play Services, Android Things attempted to lower the barrier to entry for developers coming from ecosystems maintained by organizations like Stack Overflow and GitHub.

Architecture

The architecture combined a stripped-down Android userspace with a hardened Linux kernel layer, device drivers for vendors such as Qualcomm, NXP Semiconductors, and MediaTek, and a runtime supporting Java and native C++ code via Android NDK. System services integrated with Google Play Services components and cloud APIs exposed through gRPC and HTTP/REST stacks, while edge machine learning capabilities were enabled by libraries like TensorFlow Lite. Hardware abstraction used boards and modules from manufacturers including Raspberry Pi Foundation, Intel, and Allwinner Technology to expose peripherals such as GPIO, I2C, and SPI managed by HAL layers familiar to developers used to Android Open Source Project patterns.

Development and APIs

Development used Android Studio with SDKs and emulators, standard Android API patterns, and platform-specific libraries for peripheral I/O and sensor access; developers relied on tools and samples from GitHub, community forums such as Stack Overflow, and documentation maintained at Google developer sites. APIs exposed included Peripheral I/O, PWM, ADC, and I2C interfaces that mirrored existing abstractions from Android Things Peripheral I/O API examples and encouraged use of libraries like OkHttp and Retrofit for network communication, and TensorFlow Lite for on-device inference. Continuous integration workflows commonly integrated with services such as Travis CI, Jenkins, and CircleCI while artifact distribution leveraged Google Play-style mechanisms adapted from Google Play Services and Firebase App Distribution for testing.

Supported Hardware and Devices

Officially supported hardware consisted of System-on-Modules and development boards from vendors including NXP Semiconductors (i.MX family), Qualcomm Snapdragon-based modules, and third-party boards like the Raspberry Pi models used by hobbyists and developers. Commercial partners such as ASUS and Lenovo produced reference designs and certified devices in categories including smart displays, kiosks, and industrial gateways sold by distributors like Arrow Electronics and Avnet. Peripheral ecosystems from suppliers like Bosch Sensortec, STMicroelectronics, and InvenSense provided sensors and actuators commonly integrated into Android Things prototypes and production devices.

Deployment and Management

Deployment workflows mirrored mobile deployment models using managed system images, over-the-air updates coordinated through services similar to Android Device Provisioning Program and Firebase Cloud Messaging, and device fleet management solutions from third parties such as AWS IoT integrators and Microsoft Azure IoT partners. Enterprises often combined Android Things images with Mobile Device Management tools from vendors like VMware (AirWatch), Microsoft (Intune), and BlackBerry to enforce policies, and used analytics platforms from Google Analytics or Firebase Analytics to monitor device telemetry in retail, industrial, and hospitality deployments.

Security and Updates

Security architecture leveraged Verified Boot, application sandboxing inherited from Android, and regular security patching coordinated by Google with chip vendors such as Qualcomm and NXP Semiconductors. Update mechanisms emphasized over-the-air delivery, cryptographic signing of images, and staged rollouts similar to practices used by Android Open Source Project and enterprise device management solutions; incident response and vulnerability disclosures followed industry norms practiced by organizations like CERT Coordination Center and Open Web Application Security Project contributors.

History and Reception

Launched publicly after iterations announced at Google I/O, Android Things received attention from hardware startups, industry analysts at firms like Gartner and IDC, and journalists at outlets such as The Verge and Wired for its promise to repurpose Android tooling for embedded markets. Over time, Google narrowed official hardware support and shifted emphasis toward commercial customers and cloud-centric IoT offerings, prompting critique from community maintainers on GitHub and commentary in publications like Ars Technica and TechCrunch about fragmentation and long-term support. The platform was eventually deprecated as strategic priorities moved toward integrations with Google Cloud Platform IoT services and partner ecosystems, while many concepts influenced subsequent initiatives from Google and other vendors in the embedded and edge-computing spaces.

Category:Google software