A search and rescue robot design based on LiDAR technology
Yichen Ding
1,3
and Fei Ye
2,4
1
Keystone Academy, Beijing, China
2
Shanghai Jiao Tong University, Shanghai, China
3
yichendingacademic@163.com
4
yefei5212022@126.com
Abstract. For a long time, search and rescue operations during natural disasters and man-made
catastrophes have been a major challenge. Due to the rapidly changing environment in disasters,
deploying rescue teams for search missions entails significant risks. With advancements in
technology, the latest innovations can be applied to search and rescue tasks to reduce these risks.
LiDAR (Light Detection and Ranging) sensors can be installed on unmanned search and rescue
vehicles to explore the space. This article utilizes solid-state LiDAR technology, along with
various algorithms like SLAM (Simultaneous Localization and Mapping) and EKF (Extended
Kalman Filter), to design a remotely controlled unmanned exploration vehicle. By capturing
point cloud data, it enables modelling and recording of indoor or outdoor spaces, allowing for
space exploration and the identification of trapped individuals and other important rescue-related
information before rescue personnel enter the premises. This significantly reduces the risks and
time involved in search and rescue operations. The prototype vehicle designed in this paper
possesses the advantages of low cost and high flexibility, making it feasible for direct
deployment after minor optimization. Finally, the author provides a summary and outlook for
this research.
Keywords: LiDAR, rescue robotics, point-cloud, EKF, SLAM.
1. Introduction
It had been long speculated that novel technologies may be incorporated into rescue missions in critical
situations such as earthquakes and tsunamis. In major earthquakes, it is possible to identify opportunities
in using autonomous robotic exploration vehicles in exploring, charting, and modelling the interior
spaces of collapsed buildings, or even to lo cate survivors, increasing the efficiency and safety of the
rescue mission, which in turn improves the chances of survival and the safety of the emergency response
personnel. This is a major improvement over the previous system of personnel plus rescue dogs since it
decreases the risks involved in a fully manual extraction operation. A significant example of this may
be seen in the Wenchuan earthquake of 2008, which measured 8.0 on the Richter Scale and caused major
damage to structures in the vicinity of ground zero. According to Hakami et al. [1], for such events,
there is a decreasing rate of survival as time spent before rescue increases, where the survival rate
quickly drops off to less than 20% or even 5-10% after 72 hours, as illustrated by Figure 1 below. This
is known as the Golden 72 Hours. Since the rate of survival drops significantly with time, it is important
to extract trapped survivors shortly after the incident to ensure a high survival rate, preferably in under
3 days. However, search teams are often overwhelmed by the number of tasks needed to be done at the
