Reading this reminded me of my robotics subject from college.
One thing I found really interesting back then was how an autonomous robot actually works. It's basically running in a continuous loop.
> Perceive the environment through sensors
> Understand what's happening
> Plan the next action
> Execute the action
> Observe the outcome
> Repeat
The intelligence isn't just the AI model. It's the entire system working together: perception, planning, control, localization, feedback, and decision making.
The challenge was never only building the robot.
The challenge was testing it.
A small change could require deploying to hardware, running experiments, collecting data, dealing with failures, and repeating the process over and over again.
What Antioch seems to be doing is simulating that entire loop.
Instead of testing autonomy on real robots first, you can model the robot, the environment, the sensors, and the agent inside a high-fidelity simulation and iterate much faster.
Pretty cool to see robotics moving closer to a software development workflow where you can test and ship ideas much faster.
Introducing Antioch Agent. For the first time, simulate the full physical AI stack in a closed agentic loop, entirely from the browser.
Onboard your robot, build high-fidelity simulation scenes, and test in hours what takes weeks in the field.
Develop physical autonomy at the speed of software, with Antioch Agent.