In collaboration with Maum.ai
Results achieved five months after MOU signing
Manure removal completed without driver intervention
Doosan Bobcat announced Monday that it has successfully completed a field demonstration of autonomous construction equipment technology developed in collaboration with AI solutions company Maum.ai.
The two companies signed an MOU in early June to verify the feasibility of autonomous operation technology. Within about five months, they achieved autonomous operation of compact construction equipment in a real working environment.
The demonstration used vision-language-action (VLA)-based AI technology, which processes input from cameras and other sensors to perceive the surrounding environment and understand the work situation, then translates that understanding into actual equipment movements.
Unlike conventional automation that follows preset rules or human commands, the system recognizes and assesses the work environment and conditions before acting independently.
Doosan Bobcat selected livestock barn manure removal as the demonstration task. Livestock farms face labor shortages, and manure handling in particular is work that workers tend to avoid.
A Doosan Bobcat skid loader carried out manure transport and cleaning operations at a cattle barn in Pyeongtaek, Gyeonggi Province, without driver intervention.
The loader repeatedly scooped manure with its bucket, transported it to a designated location and loaded it onto a trailer standing 2.4 meters high. It also autonomously mapped a cleaning path after detecting scattered manure on the floor, then advanced with the bucket pressed close to the ground to sweep up the waste and pile it at the designated spot, performing the sequence continuously.
Park Hyun-chul, executive vice president and Chief Strategy Officer at Doosan Bobcat, said the company will "advance autonomous operation and unmanned technologies by expanding collaboration across global research and development divisions and with diverse domestic and international partners, and develop them into practical solutions that enhance customer productivity and operational efficiency."
Choi Hong-seop, chief technology officer and CEO of Maum.ai, said the demonstration was significant in that "Maum.ai refined its proprietary physical AI model to suit the working conditions of a livestock barn, enabling precise control of construction equipment and improved autonomous operation performance."
keg@heraldcorp.com
