Key points are not available for this paper at this time.
4D printing enables the creation of adaptive and reconfigurable devices by combining additive manufacturing with smart materials. This integration introduces challenges in designing printable, responsive materials and structures. Current research focuses on improving the responsiveness and mechanical performance of smart materials, but incremental advances often lack sufficient feedback for achieving specific properties, shapes, and performance targets. Inverse design has emerged as a strategy for determining material compositions and structural configurations to meet desired outputs, but its application remains limited to simple structures. Accelerating material and structural discovery is crucial for advancing 4D printing. Artificial intelligence (AI), especially machine learning (ML), offers promising solutions to address the complexity of 4D printing design. However, conventional AI approaches often lack logical reasoning, explainability, and interpretability. This review paper highlights recent achievements and challenges in 4D printing design and introduces neuro-symbolic AI as a promising approach. By combining ML's learning capabilities with the logical reasoning and semantic understanding of symbolic AI, this approach can enhance the exploration of advanced active materials and structures. The insights provided aim to guide future research toward optimizing 4D printing for broader applications and enhanced performance. • Synthesis of recent achievements and challenges in design for 4D printing. • Relevance of neuro-symbolic artificial intelligence beyond machine learning techniques. • Accelerated development of next-generation of smart materials and structures.
Bougzime et al. (Wed,) studied this question.