As a quadruped robot, Spot can operate virtually anywhere a person can—going up and down stairs, indoors or outdoors, and through cramped or confined spaces. While this mobility can be impressive on its own, its real value is in automating sensing and inspection in areas inaccessible with wheeled or tracked robots and drones. You can send Spot into areas designed for people, that are unstructured, or even hazardous, to collect the data you need to understand what's going on in your facility or at your jobsite.
But how do you collect meaningful data and ensure it results in actionable insights? Using the Spot API you can develop an end-to-end solution, customised to the exact data, processes, and analysis that make sense for your operations. In addition to Spot’s out-of-the-box capabilities, the API allows you to add third-party payloads like cameras and sensors, autonomously trigger them during Autowalk missions, and integrate an ecosystem of software capabilities.
The API helps you reach data-driven results with a clear view into what matters most at your facility.
What can you do with the Spot API?
The goal of the Spot API is to enable developers and end users to expand on Spot’s existing autonomy, navigation, and data collection capabilities with a robust SDK and application ecosystem. You can think of the API in a number of layers, depending on what services you want to access or customise.
- Base Layer: The base layer is the lowest layer of the API, focused on getting network connections to the Spot API and authenticating credentials, building on existing industry standards. Our Hello Spot programming example demonstrates how to initialise the SDK to talk to the robot and start a basic sequence of actions.
- Movement API: The movement API provides all the capabilities you need to gather the information about Spot’s state and environment necessary for you to provide custom movement commands in a third-party system—for example, implementing gesture control.
- Autonomy Layer: The autonomy API includes the capabilities you would need to build end-to-end autonomous applications, navigation solutions for complex jobsites, and perform missions without a human in the loop.
- Data API: The data API is focused on integrating your own data acquisition and analysis tools on top of Spot’s existing capabilities. For example, you could integrate a third-party camera or implement a computer vision model.
Fundamentally, the API makes it easy to customise Spot, turning it into a robust platform that meets your unique requirements for safe, effective automation in people-centric environments.
How to go from data capture to data insights
While the API opens up a world of possible customisations, one of the most common applications is to integrate your data—to capture visual, audio, and other data, to add analysis tools, and to export the data and put it to use across your organisations. And that's where the Data API comes in.
The first step is capturing the data you need. Within the API, the data acquisition service enables you to do two key things. First, you can introduce new types of sensors to Spot and, secondly, you can integrate them with our existing controls in Spot’s tablet. For example, you can add your preferred sensors—from cameras, to gas sensors, to acoustic imagers—and then trigger scans or measurements during an Autowalk mission or while teleoperating the robot.
Once you've acquired relevant data, the next phase is analysis. The Network Compute Bridge allows you to quickly and easily implement machine learning and computer vision models on the robot and view the results on the Spot tablet. You can also feed data directly into other platforms—for example, an enterprise asset management (EAM) system—to either perform the data analysis or trigger automated actions based on the results; for example, you could generate a work order based on a thermal anomaly. The end result is an end-to-end application customised for your specific data needs.
Source: Boston Dynamics