**Eliminate Offline Slicing: Adaptive Toolpath Software Delivers Autonomous Robotic Additive Manufacturing**

This technology is a real-time software framework that enables robots to dynamically generate and adapt obstacle-avoiding toolpaths for additive manufacturing, ensuring efficient, smooth, and collision-free material extrusion in changing environments.

Background

Robotic additive manufacturing (AM) is transforming the construction industry by enabling the automated extrusion of materials like concrete for terrestrial and extraterrestrial infrastructure. In these complex scenarios, autonomous robots are deployed to build large-scale structures. However, these operational workspaces are rarely predictable and are often cluttered with dynamic obstacles. Consequently, there is a critical need for highly adaptive manufacturing systems that can seamlessly navigate around physical barriers while continuously depositing material. Autonomous construction requires systems capable of responding to real-time environmental changes to ensure structural integrity and operational efficiency.

Despite this need for adaptability, current robotic AM workflows rely heavily on conventional slicing software that generates static, offline toolpaths, such as standard G-code. These traditional methods require the entire deposition route to be pre-calculated before manufacturing begins. Because these offline approaches lack real-time environmental awareness, they are fundamentally incapable of adjusting to dynamic, obstacle-filled workspaces. If an unexpected barrier appears during extrusion, the pre-programmed robot cannot recalculate its trajectory, leading to potential collisions or complete process failure. This reliance on rigid, predetermined pathing severely limits the deployment of autonomous construction robots in unpredictable real-world environments.

Technology

The environment-aware Path Generation Framework is a software solution for robotic additive manufacturing that dynamically generates collision-free toolpaths in real time. Requiring only target boundary points for open or closed structures, the Python-based system utilizes optimized search-based algorithms, such as Dijkstra and A*, to calculate optimal routes around obstacles. It seamlessly interfaces with industrial robotic manipulators and perception systems tailored for concrete extrusion. To ensure structural viability, the framework evaluates toolpaths using precise quantitative metrics, including run time, path roughness, number of turns, offset, and root mean square error.

This technology is differentiated by eliminating the limitations of traditional, offline CAD-to-G-code slicing software, which cannot adapt to unpredictable environments. By enabling real-time path planning, the framework allows autonomous robots to navigate complex, obstacle-filled workspaces during terrestrial or extraterrestrial construction. The system is uniquely optimized for physical manufacturing constraints, tuning its path resolution to exactly match the width of extruded concrete filament. When integrated with closed-loop extrusion control and computer vision, this adaptive solution compensates for material uncertainties and environmental changes, enabling fully autonomous infrastructure development without relying on pre-programmed, static toolpaths.

Benefits

  • Enables real-time, dynamic generation of obstacle-avoiding toolpaths, allowing robots to instantly adapt to environmental changes.
  • Streamlines the manufacturing workflow by eliminating the need for traditional, static offline CAD-to-G-code slicing.
  • Ensures safe, collision-free navigation in complex, obstacle-dense environments for both terrestrial and extraterrestrial construction.
  • Optimizes structural quality and computational efficiency by automatically evaluating paths for smoothness, minimal deviation, and reduced run times.
  • Guarantees physically feasible material deposition by matching the algorithm’s path resolution directly to the physical width of the extruded material.
  • Facilitates fully autonomous construction by seamlessly integrating with real-time feedback mechanisms, such as computer vision and closed-loop extrusion control.

Applications

  • Autonomous robotic construction sites
  • Extraterrestrial habitat 3D printing
  • Real-time concrete extrusion printing
  • Military field infrastructure development
  • Adaptive industrial 3D printing

Additional info

This Python-based software framework dynamically generates real-time, obstacle-avoiding toolpaths for robotic additive manufacturing, replacing offline CAD-to-G-code slicing. Utilizing Dijkstra, A*, PRM, and RRT algorithms, the system calculates collision-free paths between target boundary points. It integrates with robotic manipulators and perception systems, evaluating path quality using metrics like run time, roughness, and root mean square error to ensure structural and computational efficiency during material extrusion.