The versatile autonomous excavator system, created by the University of Maryland in partnership with Baidu Science Corporation (RAL), is capable of unloading goods remotely without the assistance of workers. Researchers say that this technology is as close as possible to the actions of workers (people) who are professionally engaged in, for example, operator activities.
Often, an autonomous excavator system takes into account the sequence of actions online for the sake of correct understanding, as well as management and design of ideas. A lot of special cameras and sensors have been created for perfectly correct perception of the surrounding space in the form of 3D, this also includes various sensors, systems, and even LiDAR technology. Thanks to all this set, the system will be able to adequately perceive the surrounding environment, find the necessary things, recognize them and distribute them by sight. Also, for successful adaptation to the external environment, clear vision, and compliance with all the required algorithms, the technology uses a neural network. It is these factors that make it possible to use such a system on absolutely different machines.
A development corporation and a group of scientists began researching the performance of AES. Namely, a number of experiments were carried out to prove the feasibility and effectiveness of using the presented technology. The system has proven its long-term efficiency at a landfill with waste, having managed to work continuously without outside interference for more than one day. It is a very good indicator and relevant way to clean up waste because such work is very toxic and dangerous to human health.
Also, the system has been tested in various temperature modes. It is resistant to frost and temperature extremes and protected from moisture. The productivity of small machines is equivalent to the work of one employee.
It was calculated that AES can function for a fairly long time without interruptions and third-party interventions. What can not be said about mentioning the human factor with force majeure and fatigue.
Baidu Research RAL plans to further study and improve the operation of autonomous excavator systems. The system still has to go through several tests and experiments, changes in the algorithm of work, and other things. All additional materials about the system can be seen in one of the headings of the popular science journal Science Robotics.
It is worth mentioning that excavators are the most necessary machines for construction, repair work, mining, and other activities. According to research data, in 2018, the global market turnover in this area was $ 4.12 billion, while in 2026, this figure could reach $ 63.14 billion.
Baidu is introducing its AES development into the world's largest corporations that develop technologies to improve the quality and efficiency of construction machines using innovative techniques, including AES.