MIT's robotic forklift

A semi-autonomous forklift may handle movement of supplies in a hostile military war zone if a research project achieves its goals.

In the current effort, researchers are using an internal-combustion pneumatic-tyre Toyota 8-Series forklift in outdoor tests on the Cambridge campus of the Massachusetts Institute of Technology (MIT). The forklift has a 4,000-pound (1,800kg) lifting capacity.

Operated remotely, the conceptual robotic forklift is intended to transport pallet-loaded supplies outdoors on uneven terrain to depot storage locations or onto a truck for deployment. Now, in Iraq, for instance, operators at military outposts frequently abandon a forklift when nearby combat conditions elevate.

The research team hopes to have a functional prototype ready in a few months.

MIT says the research began with a small test platform that was rigged with forklift tines, sensors and computers for indoor tests.

The work is part of several projects at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) focused on "the development of situational awareness for machines", explains Seth Teller, professor of computer science and engineering and project leader. Teller co-heads CSAIL's robotics, vision and sensor networks group.

Teller's team is in the initial stages of funding and, currently, is negotiating follow-on funding for the project with the logistics innovation agency of the US Army.

For now, guidance from a human supervisor determines which pallets to lift and where to move them.

A supervisor's tablet computer, which is wirelessly linked to the forklift, displays the view from the forklift's forward-looking video camera. Directions come from stylus gestures on the image and spoken commands. The machine can revert to a conventional manned forklift as needed.

Teller says that situational awareness involves the use of sensing, motion, inference and memory to acquire "a model of the spatial layout of the world and its contents, to allow us to plan and move purposefully in the world". Humans develop these internal maps of their surroundings without even thinking about it, Teller notes, but "machines can't yet do it automatically".

The forklift project has involved about 30 MIT-connected faculty, staff and students along with support from the MIT-managed Lincoln Laboratory, the independent non-profit Charles Stark Draper Laboratory Inc and BAE Systems plc.

In an earlier project, MIT created the forklift-concept computer code for an autonomous vehicle in the grand challenge auto race organised by the Defence Advanced Research Projects Agency of the US Department of Defence.