Research and Development Status of Rescue Robot at Home and Abroad

1 Introduction

After a disaster such as an earthquake, fire, or mine disaster, search for survivors in the ruins. Giving the necessary medical assistance and saving trapped people as soon as possible is an urgent task facing rescue workers. Practical experience shows that survivors trapped in the rubble after 48 hours are less likely to survive.

Due to the complicated situation on the scene of the disaster, the safety of the rescue personnel cannot be guaranteed, and the narrow space formed in the rubble makes rescue workers and even rescue dogs unable to enter. Disaster-relief robots can solve these problems well.

2. Research status at home and abroad

In the past decade or so, especially after the “ 911 ” incident, the United States, Japan, and other Western developed countries have done a lot of work on the research of rescue robots such as earthquakes and fires. Different rescue robots using traction and sports can be mainly divided into the following categories:

1 、Crawler rescue robot

2. Deformable ( multi-state ) rescue robot

3 , bionic rescue robot

2.1 Tracked rescue robot

Tracked robots are developed on the basis of traditional wheeled mobile robots to meet the needs of military reconnaissance and the removal of dangerous objects. They were mainly developed to meet the needs of the military. The volume is generally too large to be suitable for searching for survivors in the narrow space of collapsed buildings.

2.2 Variability (Polymorphism) Rescue Robot

In order to enter a confined space, the size of the robot is required to be as small as possible, but the search field is limited due to its small size. In order to resolve this contradiction, in recent years, based on the traditional traction rescue robot platform, a variable-form crawler type has been developed. Polymorphic rescue robot.

2.3 Bionic Rescue Robot

Although the crawler-type deformable multi-state robot can change its shape and size according to the size of the search space, the volume cannot be made small by the limitation of the driving method. In order to meet the need for more narrow space search, people have developed a variety of smaller bionic robots based on ecological principles, among which serpentine robots are an important category.

2.4 Current Research Situation in China

In 2005 , the China-Japan Rescue and Safety Robot Technology Research Center jointly established by the Shenyang Institute of Automation, Chinese Academy of Sciences and the Japan International Rescue System Research Institute was unveiled in Shenyang. This signifies that China’s search and rescue robot research has entered a period of rapid development.

In 2001 , the National University of Defense Technology developed a snake-like robot 1.2 meters long , 0.06 meters in diameter , and 1.8 kilograms in weight . It can twist like a snake and can advance, retreat, turn and accelerate. Its maximum speed is up to one minute. 20 meters , the head is the robot's control center.

In June 2006, the reliability of disaster relief robot developed by the Institute of China University of Mining and China's first coal mine rescue robots (prototype) was born. This search-and-rescue robot adopts a walking control method combining self-avoidance obstacle avoidance with remote guidance. It can penetrate deep into the accident mine and detect information such as fire temperature, gas concentration, disaster scenes, and call for help in front. At the same time, the search-and-rescue robots carry emergency medicines, foods, life support fluids, and simple self-help tools.

3 , rescue robot key technical problems

3.1 Mobility/mechanism

Mobility is the decisive factor in the rescue robot's rescue work. The robot mobile platform should be able to flexibly navigate through narrow spaces in harsh ruin environments, be able to climb over obstacles, climb stairs, cross muddy roads, etc., and the movement of robots should not affect the surrounding unstable structures, so as to avoid secondary occurrences. Collapse or explosion. In addition, the robot should also have the ability to adapt to harsh environments, with the ability to waterproof, high temperature resistance.

3.2 Sensing detection device

The main task of the rescue robot is to achieve its own navigation through sensors, access to environmental information, and search for survivors. Due to the complexity and uncertainty of the disaster scene environment, traditional navigation algorithms that are more mature in the indoor structured environment can not meet the requirements of rescue work. Traditional sonar, laser range finder, etc. are in the environment full of smoke and dust. It is also difficult to achieve the desired results. At present, the rescue robot mainly uses manual control to realize the navigation of the robot.

3.3 Man-machine communication

There are currently used wireless communication and cable in two ways.

The cable method can stably and reliably transmit information between the robot and the operator. However, there are also certain problems in the cable method. As the search scope of the robot is deepened, the cable is easily entangled and affects the mobility of the robot. The development of a flexible cable winding device is the key to solve the current wired communication mode robot communication problems.

The stability of wireless communication methods is difficult to guarantee. Even in the band with the best penetration performance, the communication cannot be performed normally due to the influence of bandwidth and various types of interference. Rescue work on the “ 911 ” incident proved that wireless robots could not communicate normally for more than 25 % of the time. The stable and reliable communication method is one of the key issues that need to be solved in the field of rescue robots.

3.4 Sensor fusion

Due to the complexity of the rescue site environment, the sensor data processing algorithms in the traditional indoor structured environment can not meet the needs of rescue work. For example, the detection of survivors through video images makes it difficult to recognize due to the effects of dust, smoke, etc., and it is difficult to identify the direction of the survivor by detecting the direction of the sound and also due to the influence of the scene noise. Therefore, in order to complete the search and find survivors, a more effective identification algorithm must be studied through the fusion of various sensor data.

The information in this article comes from the Internet and was reorganized and edited by China Rescue Equipment Network.