The human arsenal is known for going deeper than any recognizable boundary, but at the same time, it is still yet to see anything more significant than that desire of ours to improve at a consistent pace. We say this because the stated desire has already fetched the world some huge milestones, with technology appearing as a rather unique member of the group. The reason why technology’s credentials are so anomalous is purposed around its skill-set, which was unprecedented enough to realize all the possibilities for us that we couldn’t have imagined otherwise. Nevertheless, a closer look should be able to reveal how the whole runner was also very much inspired by the way we applied those skills across a real world environment. The latter component was, in fact, what gave the creation a spectrum-wide presence and made it the ultimate centerpiece of every horizon. Now, having such a powerful tool run the show did expand our experience in many different directions, but even after reaching so far ahead, this prodigious concept called technology will somehow keep on delivering the right goods. The same has grown to become a lot more evident in recent times, and assuming one new discovery pans out just like we envision, it will only propel that trend towards greater heights over the near future and beyond.

The researching team at University of Bristol has successfully developed a new Bi-Touch system, which is designed to let robots leverage a digital helper to sense what manual actions are required in a given situation. Even though bimanual manipulation with tactile feedback is something well known for its potential in regards to realizing human-like dexterity among robots, the concept remains largely untapped than single-arm settings. This is, more or less, due to limited availability of suitable hardware, along with the sheer complexity of designing effective controllers for tasks with big state-action spaces. The team behind our new development overcame that by building up a virtual world which contained two robot arms boasting smart tactile sensors. Next up, they conceived reward functions and a goal-update mechanism for the purpose of encouraging the robot agents to learn to achieve the bimanual tasks. Talk about how the robot would get the stated bimanual skills, it banked upon Deep Reinforcement Learning, one of the most advanced learning techniques in the field of robotics meant to provide the required training through trial and error.

“Our Bi-Touch system showcases a promising approach with affordable software and hardware for learning bimanual behaviors with touch in simulation, which can be directly applied to the real world. Our developed tactile dual-arm robot simulation allows further research on more different tasks as the code will be open-source, which is ideal for developing other downstream tasks,” said Professor Nathan Lepora, co-author on this study.

In a more practical sense, the robot learns to make the right decision using an approach where it tries out a host of different behaviors to achieve the eventual task. These behaviors can materialize in the context of utterly simple activities like lifting up objects without dropping or breaking them. To make the robot retain an effective behavior, it is also provided a reward on every successful attempt. Conversely, as a way to discourage an ineffective behavior, we follow it up with some sort of a punishment. Over time, the latter actions are eliminated and a rather dexterous robot is born.

Coming back to University of Bristol’s latest brainchild, the researchers have already tested it across different situations, and if we put our stock in the available details, the dual arm robot was able to safely lift items as fragile as a single Pringle crisp.

“With our Bi-Touch system, we can easily train AI agents in a virtual world within a couple of hours to achieve bimanual tasks that are tailored towards the touch. And more importantly, we can directly apply these agents from the virtual world to the real world without further training,” said Yijiong Linfrom, the Faculty of Engineering and lead author on this study.