In a world first, an excavator has performed material loading tasks for 24 hours without any human intervention.
Researchers from Baidu Research Robotics, Auto-Driving Lab – RAL – and the University of Maryland have developed an autonomous excavator system – AES.
Academic journal Science Robotics published an article following the success of the trial, with report authors saying the excavator performed the task efficiently, operating “robustly” in challenging scenarios.
“This work presents an efficient, robust and general autonomous system architecture that enables excavators of various sizes to perform material loading tasks in the real world autonomously,” says Dr Liangjun Zhang, corresponding author and the Head of Baidu Research Robotics and Auto-Driving Lab.
The AES has been tested in real- life scenarios, operating for over 24 hours. Baidu says it can help companies struggling to find skilled heavy machinery operators as well as reducing the risk of having human operators working in hazardous or toxic environments.
AES uses real-time algorithms for perception, planning and control alongside a new architecture to operate the machine. Multiple sensors – including LiDAR (light detection and ranging to build pictures from laser pulses), cameras, and proprioceptive sensors – perceive the 3D environment around the machine and identify target materials. Baidu says that AES can be used on excavators of all sizes.
To evaluate the efficiency and robustness of AES, it was tested at a waste disposal site, where it operated continuously for more than 24 hours without any human intervention. AES has also been tested in winter weather conditions, where vaporisation can pose a threat towards the sensing performance of LiDAR.
The compact excavator extracted dry and wet materials at a rate of 67.1 cubic metres per hour, which is in line with the performance of a traditional human operator.
“AES performs consistently and reliably over a long time, while the performance of human operators can be uncertain,” Zhang says.
“This represents a key step moving towards deploying robots with long operating periods, even in uncontrolled indoor and outdoor environments,” says Distinguished University Professor of Computer Science and Electrical and Computer Engineering at the University of Maryland Dinesh Manocha.
Going forward, Baidu Research RAL will continue to refine core modules of AES and further explore scenarios where extreme weather or environmental conditions may be present.