面向先验知识与PSO的家电家具布局优化算法研究

doi:10.19784/j.cnki.issn1672-0172.2025.99.018

摘要/Abstract

摘要： 随着建筑行业的数字化转型和智能家居场景的快速发展,对家电家具的自动化布局方法提出了更高要求。为满足智能空间中家具与家电高效、合理摆放的需求,提出一种结合先验知识与粒子群算法（PSO）的室内布局优化方法。首先,基于空间语义规则对主要家电与家具对象进行先验分组,通过规则系统实现初始合理放置。随后构建考虑家具间关系与户型边界条件的多目标代价函数,利用PSO对整个空间中的对象进行迭代建模与布局优化。该方法引入了户型结构信息和商品组合知识,提升了初始布局的准确性与收敛效率,同时通过优化算法兼顾了单个家具对象的位置最优性与整体空间的协调性。实验结果表明,该方法生成的布局方案表现良好,可与人工设计方案媲美。研究表明,该方法属于人工智能算法在家电、家具布局优化中的典型应用,具有良好的智能家居落地前景。

关键词: 室内设计, 自动布局, 先验知识, PSO优化

Abstract: As the construction industry undergoes digital transformation and smart home environments continue to evolve, the demand for automated layout solutions for home appliances and furniture has grown significantly. Introduces an interior layout optimization method that integrates prior knowledge with Particle Swarm Optimization (PSO) to enable efficient and intelligent spatial arrangement. Initially, major furniture and appliance objects are grouped based on spatial semantic rules, allowing a rule-driven system to generate a reasonable initial placement. A multi-objective cost function is then formulated, incorporating both object-to-object relationships and spatial constraints derived from floor plan structures. PSO is applied to iteratively optimize the layout across the entire space. By leveraging structural layout information and predefined item groupings, the method enhances both the accuracy of the initial arrangement and the convergence speed of the optimization process. The approach balances individual object placement with overall spatial harmony. Experimental results show that the generated layouts are both practical and esthetically comparable to manually designed configurations. This work exemplifies the application of artificial intelligence algorithms in the optimization of furniture and appliance layouts, and holds strong potential for deployment in smart home design systems.

Key words: Interior design, Automatic layout, Prior knowledge, Particle swarm optimization

中图分类号:

TM925

陈一帆, 尹德帅, 崔振, 梁旭亮, 陶海峰. 面向先验知识与PSO的家电家具布局优化算法研究[J]. 家电科技, 2025, 0(zk): 86-89.

CHEN Yifan, YIN Deshuai, CUI Zhen, LIANG Xuliang, TAO Haifeng. A study on layout optimization algorithm for home appliances and furniture based on prior knowledge and PSO[J]. Journal of Appliance Science & Technology, 2025, 0(zk): 86-89.

/ 推荐

参考文献

[1] Yoshiaki Akazawa, Yoshihiro Okada and Koichi Niijima. Automatic 3D scene generation based on contact constraints[A]//In Proc. Conf. on Computer Graphics and Artificial Intelligence[C] , 2005: 593-598.
[2] T. Germer and M. Schwarz. Procedural Arrangement of Furniture for Real-Time Walkthroughs[J]. Computer Graphics Forum, 2009, 28(8): 2068-2078.
[3] Tim Tutenel, Rafael Bidarra, Ruben M.Smelik, et al. Rule-based layout solving and its application to procedural interior generation[C]. In CASA’09 Workshop on 3D Advanced Media in Gaming and Simulation, 2009: 15-24.
[4] R. Akase and Y. Okada. Automatic 3D Furniture Layout Based on Interactive Evolutionary Computation[C]. Seventh International Conference on Complex, Intelligent, and Software Intensive Systems, Taichung, Taiwan, 726-731.
[5] Paul Merrell, Eric Schkufza, Zeyang Li and et al. Interactive Furniture Layout Using Interior Design Guidelines[EB/OL]. ACM Trans. Graph. 30, 4, Article 87, 10 pages, 2011.
[6] Stephane Sanchez, Olivier Le Roux, Herv´e Luga et al. Constraint-Based 3D-Object Layout using a Genetic Algorithm[C]. In International Conference on Computer Graphics and Artificial Intelligence, 2003.
[7] Manyi Li, Akshay Gadi Patil, Kai Xu, et al.GRAINS: Generative Recursive Autoencoders for INdoor Scenes[EB/OL]. ACM Trans. Graph. 38, 2, Article 12, 16 pages. 2019.
[8] Zhang Zaiwei, Zhenpei Yang, Chongyang Ma, et al.Deep Generative Modeling for Scene Synthesis via Hybrid Representations[J]. ACM Transactions on Graphics (TOG) 2018, 39: 1-21.
[9] 田云龙, 王统帅, 牛丽. 智能家居领域利用AIGC大模型实现全屋智慧场景自编排交互体验的系统和方法[J]. 家电科技, 2025(04): 443-447.
[10] Yi-Ting Yeh, Lingfeng Yang, Matthew Watson, et al.Synthesizing Open Worlds with Constraints Using Locally Annealed Reversible Jump MCMC[EB/OL]. ACM Trans.Graph. 31, 2012.
[11] Matthew Fisher, Daniel Ritchie, Manolis Savva, et al.Example-based Synthesis of 3D Object Arrangements[EB/OL]. ACM Trans. Graph. 31, 6, Article 135, 11 pages. 2012.
[12] Paul Guerrero, Stefan Jeschke,Michael Wimmer and Peter Wonka. Learning Shape Placements by Example[EB/OL]. ACM Trans. Graph. 34, 4, Article 108 , 13 pages. 2015.
[13] Xi Zhao, Ruizhen Hu, Paul Guerrero, et al.Relationship Templates for Creating Scene Variations[EB/OL]. ACM Trans. Graph. 35, 6, Article 207, 13 pages. 2016.
[14] 刘米兰, 蒋浩, 许威威, 等. 基于案例库的室内区域自动布局方法[J]. 计算机辅助设计与图形学学报, 2017, 29(01): 90-100.
[15] 宋佩华, 贾金原. 基于案例推理和协同过滤的室内场景生成算法[J]. 系统仿真学报, 2019(02): 263-274.
[16] Matthew Fisher, Manolis Savva, Yangyan Li, et al.Activity-centric Scene Synthesis for Functional 3D Scene Modeling[EB/OL]. ACM Trans. Graph. 34, 6, Article 179, 13 pages. 2015.
[17] Rui Ma, Honghua Li, Changqing Zou, et al.Action-driven 3D Indoor Scene Evolution[EB/OL]. ACM Trans. Graph. 35, 6, Article 173, 13 pages. 2016.
[18] Lap-Fai Yu, Sai-Kit Yeung, Chi-Keung Tang, et al.Make It Home: Automatic Optimization of Furniture Arrangement[EB/OL]. ACM Trans. Graph. 30, 4, Article 86, 12 pages. 2011.
[19] Peter Kán, Hannes Kaufmann.Automated interior design using a genetic algorithm[A]//In Proceedings of the 23rd ACM Symposium on Virtual Reality Software and Technology (VRST ‘17)[C]. Association for Computing Machinery, New York, NY, USA, 2017: 1-10.
[20] 刘米兰, 蒋浩, 毛天露, 等. 基于复合模型的家居场景生成方法[J]. 计算机学报, 2017, 40(11): 2533-2545.
[21] David Kent Ballast.Interior Design Reference Manual: Everything You Need to Know to Pass the NCIDQ Exam[M]. Professional Publications Incorporated, 2013.
[22] M. Mitton and C. Nystuen. Residential Interior Design: A Guide To Planning Spaces[EB/OL]. Wiley, 2011.
[23] L. O’Shea, C. Grimley, M. Love.The Interior Design Reference & Specification Book: Everything Interior Designers Need to Know Every Day[M]. Rockport Publishers, 2013.
[24] J. Kennedy, R. Eberhart.Particle swarm optimization[A]//In Proceedings of the IEEE International Conference on Neural Networks[C]. 1995: 1942-1948.