Aerial Additive Building Manufacturing: Distributed Unmanned Aerial Systems for in-situ manufacturing of the built environment

空中增材建筑制造：用于建筑环境现场制造的分布式无人机系统

• 批准号: EP/N018494/1
• 负责人: Mirko Kovac
• 金额: $ 295.3万
• 依托单位: Imperial College London
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2016
• 资助国家: 英国
• 起止时间: 2016 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FN018494%2F1
• 关键词: Aerial; Additive; Building; Manufacturing; Distributed

Additive Building Manufacturing (ABM) is transforming the construction industry through the 3D printing of buildings and building components. A number of countries are now demonstrating ABM can substantially reduce construction time, material and transport costs, improve worker safety standards and alleviate construction's impact on urban traffic congestion and the environment. ABM also provides geometrical variety at no additional cost. In contrast to most manufacturing sectors, variety is a necessity within construction to satisfy different client requirements and adapt to unique terrain, boundary and laws governing each physical site. However, current ABM systems are difficult to deploy on construction sites due to their large size and fixed 3D Print build volumes that are not sufficiently flexible to deal with the complexities of most building scenarios, or provide adequate measures for human safety. These ABM technologies are unable to undertake maintenance and repair work, or construct buildings in many urban or elevated sites. They are also not able to be utilised for post-disaster reconstruction activities where their manufacturing speed would be of great assistance.To address this limitation, this research proposal aims to develop the world's first Aerial Additive Building Manufacturing (Aerial ABM) System consisting of a swarm of aerial robots (Unmanned Aerial Systems (UAS)) that can autonomously assess and manufacture building structures. Aerial ABM offers major improvements to human safety, speed, flexibility, and manufacturing efficiency compared to existing ABM and standard building construction technologies. We have already developed and demonstrated pilot results using UAS that can extrude 3D Print material during flight and we have developed simulation environments that allow for autonomous planning and execution of manufacturing with swarms of UAS working in collaboratively. Using the resources of the EPSRC grant, we will co-develop and demonstrate a working Aerial ABM system that will manufacture structural elements such as walls and a freeform building pavilion. This will require innovation and major technical contributions in Hardware, Autonomy as well as in Materials and Structures. Building on the consortium's world-leading expertise in these areas and support from industrial partners (Skanska, Ultimaker, BuroHappold, Dyson and BRE), we aim at delivering the following main research contributions through this grant:Aerial ABM Hardware- A novel Aerial ABM robot design with autonomous vision based stabilisation, navigation and mapping of a dynamically changing environment that is optimised for flight and 3D Printing tasks.Aerial ABM Autonomy- A framework for autonomous manufacturing that utilises swarm intelligence for collaborative robot-to-robot operations, dynamic task sharing/allocation, adaptive response to context and dynamic environment content involving functions such as new methods of collision avoidance.- Develop new modes of communication and control that enable the safe co-existence and cooperation of human workers, other robots and Aerial ABM robots on construction sites. Novel research in human-robot interaction, feedback and haptic interface functionalities will enable manufacturing flexibility suitable for construction sites that are always unique in size, shape and contextual complexity.- An integrated design and real-time structural analysis software that delivers optimal structural integrity from minimal material weight within building design strategies that leverage this free-form manufacturing process to create innovative building design possibilities.Aerial ABM Materials and Structures- Development of new high-performance 3D-printable composite material and deposition procedures for the additive manufacture (3D Printing) of free-form light-weight building structures utilising autonomous UAS.

增材建筑制造（ABM）正在通过建筑物和建筑构件的3D打印来改变建筑行业。一些国家现在正在证明，ABM可以大大减少施工时间、材料和运输成本，提高工人安全标准，减轻施工对城市交通拥堵和环境的影响。反弹道导弹还提供几何变化，没有额外的费用。与大多数制造业不同，建筑业必须多样化，以满足不同的客户需求，并适应独特的地形，边界和法律管理每个物理站点。然而，目前的ABM系统很难在建筑工地上部署，因为它们的尺寸很大，而且3D打印构建体积固定，不足以灵活地处理大多数建筑场景的复杂性，或者为人类安全提供足够的措施。这些反弹道导弹技术无法进行维护和维修工作，或在许多城市或高架场所建造建筑物。为了解决这一问题，本研究计划开发世界上第一个空中增材建筑制造（Aerial Additive Building Manufacturing，简称ABM）系统，该系统由一群空中机器人（UAS）组成，可以自主评估和制造建筑结构。与现有的ABM和标准建筑施工技术相比，空中ABM在人员安全、速度、灵活性和制造效率方面都有重大改进。我们已经开发并展示了使用UAS的试点结果，可以在飞行过程中挤出3D打印材料，我们已经开发了模拟环境，允许自主规划和执行制造，并与成群的UAS协同工作。利用EPSRC赠款的资源，我们将共同开发和演示一个工作的空中ABM系统，该系统将制造结构元素，如墙壁和自由形式的建筑展馆。这将需要在硬件、自动化以及材料和结构方面的创新和重大技术贡献。依托该联盟在这些领域的世界领先的专业知识和工业合作伙伴的支持（Skanska，Ultimaker，BuroHappold，Dyson和BRE），我们的目标是通过这笔赠款提供以下主要研究贡献：Aerial ABM Hardware-一种新颖的Aerial ABM机器人设计，具有基于自主视觉的稳定性，导航和动态变化环境的映射，针对飞行和3D打印任务进行了优化。Aerial ABM Autonomy-一种自主制造框架，利用群体智能进行机器人对机器人的协作操作，动态任务共享/分配、对上下文和动态环境内容的自适应响应，涉及诸如避免碰撞的新方法的功能。开发新的通信和控制模式，使建筑工地上的人类工人、其他机器人和空中反弹道导弹机器人能够安全共存和合作。在人机交互、反馈和触觉界面功能方面的新研究将使制造灵活性适用于尺寸、形状和环境复杂性始终独特的建筑工地。一个集成的设计和实时结构分析软件，可在建筑设计策略中以最小的材料重量提供最佳的结构完整性，利用这种自由成型制造工艺创造创新的建筑设计可能性。空中ABM材料和结构-开发用于增材制造的新型高性能3D打印复合材料和沉积程序（3D打印）使用自主UAS的自由形式的轻型建筑结构。

项目成果

Deep Neuromorphic Controller with Dynamic Topology for Aerial Robots

• DOI: 10.1109/icra48506.2021.9561729
• 发表时间: 2021-05
• 期刊: 2021 IEEE International Conference on Robotics and Automation (ICRA)
• 作者: Basaran Bahadir Kocer;Mohamad Abdul Hady;Harikumar Kandath;Mahardhika Pratama;M. Kovač

From Norm to Swarm: development of a balanced scorecard for evaluating automation in construction

从规范到集群：开发用于评估建筑自动化的平衡计分卡

• 发表时间: 2019
• 作者: Augusti Juan, I

Unmanned Aerial Sensor Placement for Cluttered Environments

• DOI: 10.1109/lra.2020.3015459
• 发表时间: 2020-10-01
• 期刊: IEEE ROBOTICS AND AUTOMATION LETTERS
• 影响因子: 5.2
• 作者: Farinha, Andre;Zufferey, Raphael;Kovac, Mirko
• 通讯作者: Kovac, Mirko

Robotics Research

• DOI: 10.1007/978-3-031-25555-7
• 发表时间: 2023
• 期刊: Robotics Research

AI reflections in 2021

2021 年人工智能反思

• DOI: 10.1038/s42256-021-00435-7
• 发表时间: 2022
• 期刊: Nature Machine Intelligence
• 影响因子: 23.8
• 作者: Buckner, Cameron;Miikkulainen, Risto;Forrest, Stephanie;Milano, Silvia;Zou, James;Prunk, Carina;Irrgang, Christopher;Cohen, I. Glenn;Su, Hao;Murphy, Robin R.
• 通讯作者: Murphy, Robin R.