The US Army is Building a Voice Assistant Named JUDI to Control Robots
The United States Army is developing a conversational intelligence platform that will let soldiers give voice commands to robotic vehicles using natural language. Instead of requiring formal commands, the Joint Understanding and Dialogue Interface, JUDI, will be able to understand and interpret intent in its orders, clarifying them with questions as needed.
The U.S. Army Combat Capabilities Development Command’s Army Research Laboratory is building JUDI in a partnership with the University of Southern California’s Institute for Creative Technologies. Their goal for JUDI is to combine an understanding of informal language with data from its sensors to grasp the context of its orders. In the robots used for testing right now, basically very advanced miniature cars JUDI will theoretically be able to take a single command like, “go to the top of the hill” and combine camera data identifying a nearby hill with its natural language processing to work out its goal and how to achieve it, with follow-up questions to the operator as needed. The final version will be able to ask follow-up questions should it get confused.
“Dialogue will be a critical capability for autonomous systems operating across multiple echelons of Multi-Domain Operations so that Soldiers across land, air, sea and information spaces can maintain situational awareness on the battlefield,” Dr. Matthew Marge, one of the researchers, said in a statement. “This technology enables a Soldier to interact with autonomous systems through bidirectional speech and dialogue in tactical operations where verbal task instructions can be used for command and control of a mobile robot. In turn, the technology gives the robot the ability to ask for clarification or provide status updates as tasks are completed. Instead of relying on pre-specified, and possibly outdated, information about a mission, dialogue enables these systems to supplement their understanding of the world by conversing with human teammates.”
According to the Army, soldiers controlling robots with voice commands to an AI will be able to handle multiple devices while being more aware of the environment around the robot than the current, joystick-based controls. The system sounds very reminiscent of consumer voice assistants like Google Assistant, Apple’s Siri, and Amazon’s Alexa, but the Army sees JUDI as unique and distinct. According to Marge, the biggest difference is that the commercial brands use the cloud and large datasets to learn and remember a very broad set of tasks. JUDI instead relies on physical data and the phrasing of its orders to specialize in a relatively small number of duties.
“Our ultimate goal is to enable Soldiers to more easily team with autonomous systems so they can more effectively and safely complete missions, especially in scenarios like reconnaissance and search-and-rescue,” Marge said. “It will be extremely gratifying to know that Soldiers can have more accessible interfaces to autonomous systems that can scale and easily adapt to mission contexts.”
Of course, any AI that can complete a scouting mission could theoretically be trained for more deadly purposes. That’s exactly what the Russian military is working on, testing its own voice assistant to install in Marker combat drones, scaled-down tanks that go onto battlefields with humans. JUDI is not for battle deployment, at least not yet. The two AI programs do share the same goal of combining physical inputs with natural language commands to carry out tasks. Russia also integrates voice assistant named Rita in the new MiG-35 fighter jets capable of offering ideas to pilots during combat. Experiments in adding voice AI to military robots are almost certainly a global phenomenon, even if other countries are keeping their efforts quieter. No military will want to bring their AI into battle while it is prone to inevitable misunderstandings far worse than merely playing the wrong song.