In an attempt to build a robot that is different from the iPad-on-a-stick form, engineers from Japan’s Keio University and the University of Tokyo have developed a prototype telepresence bot, Fusion, which is basically a robot backpack with two arms and a head. This fun, ingenious, and even cute robot is operated remotely by a human using a VR headset and controllers. According to the lead designer Yamen Saraiji, it is supposed to encourage collaboration and learning at a distance.
Saraiji’s inspiration behind developing this robotic-arms-on-a-backpack project is to explore how people may be able to work together to control one person’s body. The arms of Fusion are actually controlled remotely by another person, who’s wearing an Oculus Rift VR headset. The handheld controllers can be wielded to direct the non-human arms and connected hands. The cameras linked to the backpack ensure a good view. The robotic arms and hands can pick things up or move around the arms and hands of the human wearing the backpack.
The most interesting fact about these arms is that they can be used in different configurations — they can move freely by themselves, be controlled by the host, or they can even attach to the wearers’ arms using wrist cuffs to move the human about.
Saraiji says, “There are some kinks to be worked out first, though. Fusion is just a prototype, and it has a number of minor flaws that limit smooth operation. For a start, the arms work at a bit of lag, an expert can guide new practitioners on how to operate certain instruments or to assist them remotely without the need of their physical presence.” Or it could help with the rehabilitation process for people in physical therapy. The elderly could even use it to pass on their physical skills to the next generation.
Lots of work is being done to give humans supernumerary limbs and create surrogate telepresence robots. The backpack also includes a PC that streams data wirelessly between the robotic arm-wearer and the person controlling the limbs in VR. The PC also connects to a microcontroller, letting it know how to position the robotic arms and hands and how much torque to apply to the joints.
