Robotic search and rescue in earthquake-collapsed buildings: capabilities, evidence gaps, and a roadmap for human-multi-robot rescue
Following an earthquake, rescue robots must search for and assist trapped individuals in complex environments characterized by unstable rubble, narrow cavities, darkness, dust, and unreliable communications. This paper reviews the technical evidence for seven categories of robots within the seven-stage mission chain of search and rescue (SAR) in building collapse scenarios, covering unmanned aerial vehicles (UAVs); tracked, wheeled, and reconfigurable robots; quadrupedal robots; snake-like and continuum robots; humanoid and hybrid robots; heavy-duty engineering robots; and casualty evacuation and medical transport robots. This paper evaluates representative “platform-sensor-mission” capabilities across two independent dimensions: validation maturity (E0–E4) and relevance to collapse rescue scenarios (S0–S4). The study clearly distinguishes between signals indicating the possible presence of trapped individuals, 3D positioning information usable to guide excavation operations, and the actual execution of subsequent rescue operations. Currently, evidence of field applications is primarily concentrated on tasks such as site entry, reconnaissance, mapping, visual search, and the detection of vital signs through rubble. In contrast, evidence regarding sustained communication, supply delivery, close-range manipulation of casualties, casualty management, evacuation, and mission handoffs is significantly lacking. This leads to the identification of three major gaps: the transition from search and detection to rescue operations; the coordinated transition between different platforms; and the assurance of safe deployment. The roadmap proposed in this paper emphasizes that human-supervised robot teams should share rescue information while preserving its uncertainty; maintain communication links where feasible; hand off tasks to appropriately equipped robots; and obtain authorization before performing critical operations involving safety. End-to-end evaluations should quantify information loss during task handoffs, positioning uncertainty, resilience following team failures, and the level of protection provided to trapped individuals.
Authors
- Songhang He
- Yuqing Gao (ORCID: https://orcid.org/0000-0001-8709-8378)
- Haiying Wang (ORCID: https://orcid.org/0009-0001-0613-6924)
- Feifei Sun
- Defeng Xu
- Zhen Wang
- Xiangdian Chen
Institutions
- Tongji University (CN)
- National Earthquake Response Support Service (CN)
- Jilin Agricultural University (CN)
- Ministry of Emergency Management of the People's Republic of China (CN)
Publication Details
- Journal
- Smart Construction
- Published
- 2026-11-16
- DOI
- https://doi.org/10.55092/sc20260020
- Primary Topic
- Robotics and Automated Systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China