AWS RoboMaker features

Building and testing applications for autonomous robots is challenging, complicated and resource intensive. Traditionally, teams of developers write code to address a wide array of deployment scenarios, the code is integrated, and then the application scenarios are tested on robotics hardware in physical environments. This manual process of development and testing consumes staff time, requires expensive hardware, slows the release cycle for application updates, and cannot be easily scaled.

With simulation, developers can test applications in virtual environments, or worlds, enabling them to increase testing coverage, reduce code errors, and accelerate development velocity. One of the most important benefits of using simulation is the ability to test and train robots in a large number and diverse set of worlds. Scaling simulation unlocks the ability to conduct robust regression testing, reinforcement learning, and synthetic data generation. However, for robotics companies, building simulation worlds, managing simulation infrastructure, and scaling testing is expensive and requires specialized skills. This often limits the use of simulation among robotics companies, and limits the benefits of automated testing at scale.

AWS RoboMaker is a fully managed service that enables developers, QAs, and DevOps engineers to easily create simulation worlds and run simulation jobs without provisioning or managing any infrastructure. RoboMaker makes simulation at scale affordable and accessible to all robotics companies by providing tools for developers to test and iterate code in 3D virtual environments. The service supports large-scale and parallel simulations, and automatically scales based on the complexity of the scenarios being tested. With RoboMaker simulation, robotics companies can make robotics application testing and machine learning faster, less expensive, and more robust.

Run large-scale, parallel simulations

With the RoboMaker batch simulation API, developers can easily launch a large batch of simulations with a single API call. Running large-scale simulations through an API makes it easier to access compute power, allowing developers to increase the complexity, scale, and frequency of their tests, which speeds up the development lifecycle and makes testing more robust.

Use any simulator

RoboMaker Simulation can be used to run the robot software and simulator of choice, including ROS, custom robot applications, Gazebo, Unity, Unreal, and Nvidia Isaac-based simulations. RoboMaker enables you to run your simulations in the cloud without provisioning, configuring, or managing any infrastructure.

Build simulation worlds at scale

RoboMaker WorldForge makes it possible to automatically create hundreds of pre-defined, randomized simulation worlds that mimic real-world conditions without engineering investment or managing world generation infrastructure. Without a simulation world generator, building a simulation world is costly, time consuming, and requires specialized skills in 3D modeling and knowledge of simulation engines. WorldForge makes it easy to increase the scale, speed, and variance of simulation, improving the quality of production code and accelerating time to market. Today, WorldForge provides worlds for indoor home environments, complete with configurable floor plans and furnishings.