In-process 3D Deviation Mapping and Defect Monitoring (3D-DM2) in High Production-rate Robotic Additive Manufacturing

TL;DR

This work tackles in-process geometric monitoring for high-deposition-rate robotic additive manufacturing (HDRRAM), notably cold spray AM, by introducing 3D-DM$^2$: a real-time framework that fuses data from a multi-sensor triad of 2D laser profilers, reconstructs the evolving surface with a TSDF volumetric fusion that employs adaptive weighting, and generates 3D deviation maps by comparing against a Gaussian-derived near-net reference model. The approach yields automated detection, segmentation, and layer-wise tracking of global and local deviations, demonstrated on complex geometries (twisted tower and seven-sided shapes) with validation against post-scanned ground truth. Key contributions include multi-sensor fusion for full-view monitoring, dynamic surface reconstruction at real-time rates, a near-net reference model tailored to HDRRAM deposition, and a robust deviation-detection pipeline capable of tracking defect evolution across layers. Collectively, the framework enables early defect detection and paves the way for closed-loop process control and reduced post-processing in high-throughput CSAM manufacturing.

Abstract

Additive manufacturing (AM) is an emerging digital manufacturing technology to produce complex and freeform objects through a layer-wise deposition. High deposition rate robotic AM (HDRRAM) processes, such as cold spray additive manufacturing (CSAM), offer significantly increased build speeds by delivering large volumes of material per unit time. However, maintaining shape accuracy remains a critical challenge, particularly due to process instabilities in current open-loop systems. Detecting these deviations as they occur is essential to prevent error propagation, ensure part quality, and minimize post-processing requirements. This study presents a real-time monitoring system to acquire and reconstruct the growing part and directly compares it with a near-net reference model to detect the shape deviation during the manufacturing process. The early identification of shape inconsistencies, followed by segmenting and tracking each deviation region, paves the way for timely intervention and compensation to achieve consistent part quality.

Figures (21)

• Figure 1: Robotics Cold Spray System Equipped with a novel 3D vision system and method using multiple 2D lasers.
• Figure 2: Experimental robotics CSAM setup, multi 2D profilers sensory system and coordinate frames
• Figure 3: Multi-sensor vision system for eliminating FOV obstruction due to tool multidirectional movement in AM. (a) A stationary scanner acquires the profile of a moving substrate. (b) A single profile sensor is not able to capture multidirectional deposition. (c) Multiple arrangements of scanners can capture multi-directional deposition.
• Figure 4: 3D-DM$^2$: overall design of 3D vision system, 3D shape reconstruction, 3D deviation mapping and defects monitoring.
• Figure 5: (a) selected voxels (orange) truncated by a distance based on ray casting, (b) Action area of deposition where the material is being added for dynamic scene update, and Inactive area of deposition used for smoothing noise in the reconstructed surface.