CoHaptics: A wearable haptic robot with a 6DoF camera to enhance the safety of human-robot collaboration

The fig of collaborative mobile robots introduced successful real-world settings is gradually increasing, with astir 20,000 caller installations per year. For their usage to summation further, however, roboticists should guarantee that these robots are some businesslike and harmless to usage astir humans.

Researchers astatine Skolkovo Institute of Science and Technology (Skoltech) person precocious created CoHaptics, a strategy that merges collaborative robotics and haptics exertion to amended the information of human-robot collaboration. This system, presented successful a insubstantial pre-published connected arXiv, tin way quality users' movements utilizing hand-tracking exertion and supply tactile feedback to pass them astir imaginable collisions with robots.

"Collaborative robots are designed to beryllium harmless for humans and are equipped with force/torques sensors," Dzmitry Tsetserukou, subordinate professor, caput of the Intelligent Space Robotics Laboratory, told TechXplore. "However, they typically lone halt moving aft a collision with objects is detected, which means that they tin inactive beryllium unsafe for delicate parts of body, specified arsenic a user's fingers, wrist, cervix and head. This is 1 of the reasons wherefore they are not wide utilized for human-robot collaborative tasks, but simply regenerate concern robots successful immoderate circumstantial tasks, specified arsenic palletizing, prime inspection, assembly, etc."

In caller years, the usage of robots has go peculiarly wide successful the manufacture and work sectors. To heighten their information and efficiency, however, roboticists should devise collaboration techniques that see some the dexterity that humans are susceptible of and the spot oregon powerfulness of robots.

"For example, if you enactment successful an assembly line, wherever the elements to assemble products are dense oregon antithetic tiny parts are used, 1 collaborative robot tin enactment you successful picking up the dense pieces and bring you adjacent to the spot to assemble the parts," Miguel Altamirano Cabrera, Ph.D. pupil progressive successful the improvement of CoHaptics, told TechXplore. "However, the robot is moving unneurotic with you, and immoderate collisions tin hap erstwhile your assemblage interferes successful the robot's path."

To trim collisions betwixt robots and humans, Tsetserukou, Altamirano Cabrera and their colleagues utilized a tracking system based connected a 6 DoF camera that tin beryllium worn by users connected their wrist. This camera collects information that tin beryllium utilized by the robot to infer the determination of quality workers successful its proximity and debar collisions with them.

To measure their system, Tsetserukou, Altamirano Cabrera and their colleagues carried retired an archetypal experiment, wherever a quality and a robot completed tasks wrong a shared environment. In this experiment, they besides compared the clip it took to implicit tasks erstwhile the quality idiosyncratic was utilizing their haptic feedback-based strategy to the cognition clip erstwhile the idiosyncratic was not utilizing the system.

The researchers recovered that, successful instances wherever the robotic strategy was alert of a user's position, their strategy reduced collisions with the robot and accrued the information of the human-robot interaction. However, the robot took longer to implicit its mission, chiefly due to the fact that it deviated from its way successful bid to debar collisions with the user.

"To lick this issue, we besides provided the users with a 'warning sign' erstwhile it is adjacent to the robot, informing them astir imaginable collisions," Altamirano Cabrera "This motion was fixed by haptic stimulation connected their forearm by a sliding touching point, representing the robot's distance."

In summation to adapting a robot's way to trim collisions with humans, the researchers frankincense added a haptic diagnostic to their strategy that informs quality users astir imaginable aboriginal collisions with robots successful advance. Their presumption was that portion moving connected an assembly enactment oregon successful concern settings humans mightiness beryllium truthful focused connected the tasks astatine manus that they mightiness not beryllium paying attraction to adjacent robots.

"Luckily, we person a consciousness of touch, truthful this is the transmission done which we decided to convey the hazard of collision to humans," Tsetserukou said. "We developed a haptic show dubbed RecyGlide that generates tactile stimuli successful accordance with the region betwixt manus of a quality and manus of a robot. When humans acquisition tactile feedback, they tin determination their manus to debar collisions with the robot."

In their strategy evaluations, the researchers recovered that users responded to the haptic feedback provided by their strategy wrong an mean clip of 0.32 sec, which is accelerated capable to forestall a imaginable collision. To summation the information of robot-human collaboration further, they besides developed an algorithm called haptic imaginable field.

The haptic imaginable tract algorithm was designed to alteration the trajectory of a robot successful a mode that prevents it from colliding with humans, without interrupting its ngo and hence continuing to determination towards a people object. Through this algorithm, the CoHaptics strategy tin usage the accusation collected by the 6DoF camera worn by users connected their wrist to modify the way of robots and forestall collisions.

"Our results showed that the accusation provided to the users by haptic stimulation accrued the human-robot region by 4.1 cm (from 12.39 to 16.55), and the robot's way was reduced by 81%," Altamirano Cabrera said. "From these results, we tin reason that the implementation of haptic stimulation to pass the users astir their region to the robot successful a collaborative task could summation the information of the users,"

In the future, CoHaptics could beryllium to beryllium a highly invaluable strategy for expanding the information of workers moving alongside mobile robots successful concern settings. Meanwhile, Tsetserukou, Altamirano Cabrera and their colleagues program to devise caller approaches that could further heighten human-robot collaboration.

As portion of their existent work, the squad is moving connected gathering a 3D environment, dubbed Coverse (collaborative universe), which represents a mill wherever robots and workers tin beryllium tracked and animated successful real-time. They could yet besides make akin representations for municipality environments oregon different settings wherever robots mightiness beryllium deployed.

"The 3D practice we are creating appears successful a gradient standard from greenish (safe) to reddish (hazard)," Tsetserukou said. "When portion of the quality assemblage is successful the proximity of a robot, the adjacent idiosyncratic receives interaction feedback, to debar the imaginable injury. This 3D collaborative beingness volition perchance marque the mill and metropolis situation populated with humans and robots safer and much natural. One tin ideate the script erstwhile you volition person haptic feedback erstwhile outdoor transportation robot is approaching you from unsighted zone. On the apical of that, users could acquisition the determination of the parcel transported by mobile robot close connected their tegument successful real-time with a transportation haptic radar to conscionable the rover connected time."

---

---

More information: Miguel Altamirano Cabrera et al, CoHaptics: Development of human-robot collaborative strategy with forearm-worn haptic show to summation information successful aboriginal factories. arXiv:2109.05912v1 [cs.RO], arxiv.org/abs/2109.05912

© 2021 Science X Network