Researchers develop an autonomous robotic system designed to charge other robots as they complete their missions.
Autonomous robots are expected to be a reality in the near future. However, one of the major challenges for realizing swarms of autonomous robots is if any of them depletes its power, then it can be a serious obstacle for pedestrians and traffic.
Researchers from Skolkovo Institute of Science and Technology in Russia have recently developed a system called MobileCharger, an autonomous robotic system designed to charge other robots as they complete their missions. The idea is to reach faulty robots or robots with a depleted battery power and assist them.
“A delivery robot and MobileCharger can run as one unit until the target robot is fully charged,” Dzmitry Tsetserukou, Professor and Head of the Intelligent Space Robotics Laboratory, says. “Subsequently, MobileCharger detaches and goes to a new robot or charging station.”
The system is similar to aerial refueling, in which airplanes are fuelled during flight operation. This allows aircraft to remain in the air longer and can lead to 35–40 percent fuel savings during long-range missions.
“Instead of losing time for traveling to the charging station, delivery robots can always be engaged in the mission,” Tsetserukou said. “In the future, it will be also possible to create a MobileCharger energy harvester with an integrated solar panel. This would mean that when the battery is near exhausted, a robot can simply move to a sunny spot and position solar panel towards sunlight.”
The MobileCharger incorporates a system that can position electrodes in three-dimensions (3D) to sustain charging in situations where a target robot is not horizontally or vertically aligned to the charger.
“Computer vision systems are not as effective as tactile ones in detecting electrode misalignment in close proximity,” Tsetserukou said. “DeltaCharger has an advanced tactile perception, as it is based on high-density pressure sensors provided by Professor Hiroyuki Kajimoto from University of Electro-Communications in Tokyo, Japan.”
The researchers developed a convolutional neural network (CNN) that can evaluate the angle of misalignment between the electrodes on MobileCharger and the robot that requires charging.
“Besides the charging of mobile robots, a future application of the technology may be the charging of drones in midair, with a charging drone based on the DeltaCharger mechanism wired to the ground power station and a target UAV being recharged without landing,” Tsetserukou said. “Innovative autonomous ships and tankers with electric motors could also be charged during sailing with the proposed concept, as this would prevent them from having to enter ports for refueling. Additionally, any types of robots such as robot dog, outdoor cleaning robots, and even self-driving cars can be charged in the similar way.”
“To increase the power of the electric source, swarm of MobileCharger robots can form parallel connection with increased capacity to charge powerful vehicles or series connection to adjust output voltage to the target vehicle,” Tsetserukou said. “In the future, we can imagine that robots will not only charge but also repair the autonomous robots to make Roboverse (robotic universe) self-sustainable.”
The researchers believe that their innovation could be adapted to charge countless different robotic systems, ranging from autonomous vehicles to mobile robots.
The research appeared in the journal IEEE Robotics and Automation Letters.