Researchers Build a Dry-Stone Wall From Massive Boulders — with an Autonomous Excavator Called HEAP

Researchers at ETH Zurich have given an autonomous excavator an unusual job: building a dry-stone wall, using boulders weighing several tons each.

"Automated building processes that enable efficient in situ resource utilization can facilitate construction in remote locations while simultaneously offering a carbon-reducing alternative to commonplace building practices," the researchers explain in the abstract to their paper. "Toward these ends, we present a robotic construction pipeline that is capable of planning and building free-form stone walls and landscapes from highly heterogeneous local materials using a robotic excavator equipped with a shovel and gripper."

That excavator is HEAP, the Hydraulic Excavator for an Autonomous Purpose, a customized Menzi MUCK M545 excavator fitted with an advanced sensor package and the smarts required for fully-autonomous operation. "The machine has novel force-​controllable hydraulic cylinders in the chassis that allow it to adapt to any terrain," the team explains of the hardware. "Additionally, HEAP is equipped with sensors necessary for autonomous operation, i.e. LIDARs, IMU, GNSS and joint sensing."

HEAP's capabilities were showcased in a 2020 project which saw the machine creating a free-form embankment from dirt and sand it sourced itself from the surrounding environment. This time, though, the machine had a more challenging task: finding and manipulating rubble and boulders, some weighing several tons apiece, and building a dry-stone wall — without any human intervention.

"Our system learns from real and simulated data to facilitate the online detection and segmentation of stone instances in spatial maps, enabling robotic grasping and textured 3D scanning of individual stones and rubble elements," the researchers explain. "Given a limited inventory of these digitized stones, our geometric planning algorithm uses a combination of constrained registration and signed-distance-field classification to determine how these should be positioned toward the formation of stable and explicitly shaped structures."

To prove the concept, HEAP was given the task of building both a 33-feet free-standing dry-stone wall, with no mortar or cement, and a 215-feet retaining wall integrated with contoured terraces. To do so, the robot mapped the site, tracked and measured the stones and rubble before grasping and positioning each — estimating their weight and center of gravity to ensure stability in the finished structure.

"The work illustrates the potential of autonomous heavy construction vehicles to build adaptively," the researchers conclude, "with highly irregular, abundant, and sustainable materials that require little to no transportation and preprocessing."

The team's work has been published in the journal Science Robotics under open-access terms.