Workshops Archive

Using Persistent Digital Twins as Spatial Environments for Robotics Simulation and Deployment


Workshop: ISAV26: In Situ AI, Analysis and Visualization

Authors: Pierre Forestier (3dverse technologies inc.)

Abstract: Deploying autonomous robots into new industrial sites requires their simulation and control software to understand the environment in which they will operate. This can create duplicated engineering effort between environments prepared for robotics simulation and 3D representations of physical sites already maintained for operational applications.

In an ongoing industrial robotics project, we are exploring whether a persistent 3D digital twin can serve as a reusable spatial environment for an external robotics simulator, with the objective of reducing the time and engineering effort required to deploy autonomous robots on new customer sites.

The robotics company maintains its own simulation, perception, navigation, and control stack. Through a bidirectional interface, robot poses are sent to the 3D environment, virtual sensor viewpoints are updated, and depth information is returned to the robotics simulator. Users can simultaneously visualize the simulation within the customer site and issue commands back to the simulated robot.

Using an API-driven, web-based architecture, the two independent systems were integrated in only a few weeks. This Community Talk will share this early industrial implementation and discuss practical questions around real-time coupling, sensor feedback, latency, and the potential for persistent digital twins to become shared spatial infrastructure across simulation and operational applications.


Back to ISAV26: In Situ AI, Analysis and Visualization Archive Listing

Back to Full Workshop Archive Listing