Ashutosh Saxena, assistant professor of computer science, has been named one of eight “Innovators to Watch in 2015” by the Smithsonian Institution.
In his Robot Learning Lab, Saxena works toward a world where robots will take over menial tasks in our homes and offices, assist the disabled and work in places too dangerous or inaccessible for human beings. As the Smitsonian puts it, “unlike the novelists and screenwriters who have also dreamt this, he is actually making it happen.”
His robots have learned a large repertoire of tasks, from pouring coffee to making dessert to putting the dirty dishes in the dishwasher. They can respond to human commands to do these things, and learn new tasks from human instructors.
Unlike industrial robots, these machines are not locked into rigid patterns of movement. Through a computer science discipline known as “machine learning” the robot’s computer brain can observe many versions of the same task and see what they all have in common, building a flexible outline to follow. Teach your robot to pour a cup of coffee and it will be able to do that no matter where you put the cup.
Soon even that won’t be needed. Saxena is building “RoboBrain,” an online database of things robots need to know to do their jobs.