Autonomous industrial control systems
Train intelligent controllers with a workflow built for engineers to design autonomous control systems that can sense and adapt to changing environments.

Project Bonsai solution
Discover ways engineers can apply their subject matter expertise to accelerate the development of intelligent control systems without the need for data science skills.

Introducing Project Moab
Get details about this open source balancing robot designed to help engineers and developers learn how to build real-world autonomous control systems with Project Bonsai.

Autonomy in action: Cartpole demo
Understand how Project Bonsai solves a pole balancing problem by applying reinforcement learning using parallel executed Simulink instances for training data generation.

Create solutions to unlock new possibilities
Experience next-generation control systems that help machinery and processes adapt in real time to changing conditions by creating AI agents that operate effectively and reliability in your most dynamic environments.

Drive significant improvements in operations
Reduce inefficiency with an intelligent control system that can optimise towards multiple goals and handle changing input while driving consistency among human operators with standardised recommendations.

Focus on your most important tasks
Give operators real-time recommendations to help them monitor operations and make the right decision at the right time. Leverage intelligent control systems to minimise tedious and time-intensive tasks.
-
Simulators supported by Project Bonsai
Project Bonsai uses simulations as training environments for advanced AI-based control systems. Engineers can use existing simulation models with the service and add intelligent decision making to their real-world system.
Explore our gallery of popular simulation tools that work with Bonsai.
-
AnyLogic
AnyLogic™ is the industry leader in dynamic simulation modelling software for business applications, enabling analysts, engineers and managers to gain deeper insights into their organisations, optimise complex systems and enable reinforcement learning.
Learn about AnyLogic -
MathWorks
MathWorks is the maker of MATLAB and Simulink, mathematical computing software used by engineers, scientists, mathematicians and researchers for data analysis, control system design, multi-domain simulations and embedded systems.
Learn about MathWorks -
MuJoCo
MuJoCo stands for Multi-Joint dynamics with Contact, developed by Emo Todorov for Roboti LLC. MuJoCo is a physics engine that facilitates development in robotics, biomechanics, graphics and animation. It offers speed, accuracy and modelling power for areas where fast and accurate simulation is needed.
Learn about MuJoCo -
Petro-SIM
Petro-SIM is KBC’s process simulation software and optimisation platform for facility performance and organisational productivity. Rigorous process simulation models in an intelligent, user-friendly environment create a leading platform for upstream, midstream and downstream refinery and petrochemical processes.
Learn about Petro-SIM from KBC -
Rhinoceros 3D
Rhinoceros 3D is an accurate, free-form modeller. Rhino can create, edit, analyse, document, render, animate and translate NURBS* curves, surfaces and solids, point clouds, and polygon meshes. There are no limits on complexity, degree or size beyond those of your hardware.
Learn about Rhino 3D

White paper: AI for Industrial Application
Developing autonomous control systems can be complex, time-consuming and expensive. Learn how we are democratising development.

Q&A about building intelligent control
Microsoft engineer Kingsuk Maitra shares an overview of autonomous systems in this blog about building intelligent control systems.

Autonomous systems support
Engage with community forums, find consulting resources and explore quick-start guides to start using the Bonsai platform.

Autonomous systems at AI Lab
Explore, learn and code with examples that apply AI innovations in machine learning, machine teaching and intelligent robotics.