Speaker
Description
Four-dimensional printing extends additive manufacturing by enabling printed structures to change shape, properties, or functionality over time in response to external stimuli. Although artificial intelligence is increasingly used in additive manufacturing for material prediction, inverse design, process optimization, defect detection, and control, the extent to which these capabilities support autonomous 4D-printing systems remains unclear. This study presents a systematic evidence map of AI applications in 4D printing and evaluates their progression toward closed-loop autonomy. The reviewed literature is classified across six lifecycle stages: material discovery, structural design, process planning, fabrication monitoring, stimulus control, and transformation validation. An AI Autonomy Maturity Index is proposed to assess five essential capabilities: sensing, prediction, decision-making, control, and learning. A complementary Lifecycle Integration Breadth measure evaluates how many stages of the 4D-printing workflow are connected within each study. The framework distinguishes isolated prediction models, AI-assisted optimization, decision-support systems, partially closed-loop processes, and highly integrated autonomous platforms. Particular attention is given to real-time sensing, automatic corrective action, physical validation, continuous model updating, and reproducibility. The study provides a structured basis for identifying current technological gaps and proposes a research roadmap centered on physics-informed AI, digital twins, multimodal sensing, uncertainty-aware control, and human-supervised autonomy. The findings are intended to guide the transition from fragmented AI applications toward reliable and integrated closed-loop 4D printing.
| Topic Area | Design Optimization, Additive Manufacturing Technologies & Applications |
|---|