The United States' (US) Army is creating an innovative software that could possibly give army machinery the ability to follow verbal orders from soldiers and communicate in natural language while on the battlefield. However, currently the technology is used for search and rescue missions while its further development is underway.
This move is followed by artificial intelligence (AI) efforts made in China, crafted to outrank the US in quantum computing in a five-year plan.
JUDI Software set to Streamline Battle Comms
Army's Combat Capabilities Development Command (aka DEVCOM) researchers and the University of Southern California are working in partnership for this project. The technology is said to be called a "JUDI: a Joint Understanding and Dialogue Interface which boasts as a conversational AI system for military machines.
The innovation is unlike the chatty digital assistants in Apple's Siri or Amazon's Alexa, as it is reportedly not designed for small talk. The software is trained to respond to spoken orders and handle certain specific situations military personnel are likely to experience while in the battlegrounds.
A DEVCOM Computer Scientist, Felix Gervits specifies the difference between the JUDI software to regular chatbot assistants. He said that the JUDI software differs from the make-up of chat text-based robots and personal assistants as the new software involves task-oriented dialogue with robots that are in the physical world, using "reason over their real-time sensory perceptions."
"It contrasts directly with current text-based chatbots and intelligent personal assistants in that it involves task-oriented dialogue with robots that are situated in the physical world and reason over their real-time sensory perceptions," Gervits says.
He also said that they have embedded a statistical classification technique and "state-of-the art" natural language that helps in the efficient conversational AI communication. It is said to recognize the purpose of the conversation and decipher its underlying content.
"We employed a statistical classification technique for enabling conversational AI using state-of-the-art natural language understanding and dialogue management technologies. The statistical language classifier enables autonomous systems to interpret the intent of a soldier by recognizing the purpose of the communication and performing actions to realize the underlying intent." Gervits adds.
Related Article : Chinese AI Unicorn Giants SenseTime and Megvii Survive US Ban and Make It Washington's Entity List! | Tech Times
How Does a JUDI Software Work Today?
The system allegedly recognized speech from the dialogue of soldiers recorded in a search-and-rescue task, and was tested by transferring it to a robot deployed in a similar exercise, with the expectation it would understand and carry out verbal natural-language commands. The classifier should map recognized commands to the necessary control signals to achieve its aspired result.
The system, for example, should be able to understand the sentence, "turn 45 degrees and send a picture," and duly rotate itself to snap and transmit a photo from its attached camera.
The DEVCOM computer scientist said that the task involved the robot navigating an urban environment and communicating with a remotely-located human, calling it an urban search-and-rescue in a disaster relief scenario. He said they used a small, man-portable ground vehicle platform equipped with a LiDAR scanner and RGB camera. It was capable of carrying out movement instructions, sending pictures upon request, providing status updates, and requesting clarification for unclear instructions.
JUDI to Innovate Future Battle Comms
The team hopes to improve JUDI by mapping more natural-language commands to its machine capabilities, so that the software can learn from more examples of real-world dialogue. This will be integrated to its computer vision with natural-language processing, which includes getting robots to analyze gestures and facial expressions.
Machines employing this technology in the future will need to be equipped with a microphone to pick up speech, a computer to process received information, and a display screen. The goal is to have robots and vehicles capable of communicating fast and instinctively to further efficiently aid human activities.
The team said that while the current focus of JUDI is on search-and-rescue tasks for now, other collaborative, remote tasks such as reconnaissance, transport, and surveillance uses are utility possibilities also being explored while the software is being developed.
This article is owned by Tech Times.
Written by Gabrielle N.