Related Articles
Categorizing robots by performance fitness into the tree of robots
Robots are typically classified based on specific morphological features, like their kinematic structure. However, a complex interplay between morphology and intelligence shapes how well a robot performs processes. Just as delicate surgical procedures demand high dexterity and tactile precision, manual warehouse or construction work requires strength and endurance. These process requirements necessitate robot systems that provide a level of performance fitting the process. In this work, we introduce the tree of robots as a taxonomy to bridge the gap between morphological classification and process-based performance. It classifies robots based on their fitness to perform, for example, physical interaction processes. Using 11 industrial manipulators, we constructed the first part of the tree of robots based on a carefully deduced set of metrics reflecting fundamental robot capabilities for various industrial physical interaction processes. Through significance analysis, we identified substantial differences between the systems, grouping them via an expectation-maximization algorithm to create a fitness-based robot classification that is open for contributions and accessible.
User-specified inverse kinematics taught in virtual reality reduce time and effort to hand-guide redundant surgical robots
Medical robots should not collide with close by obstacles during medical procedures, such as lamps, screens, or medical personnel. Redundant robots have more degrees of freedom than needed for moving endoscopic tools during surgery and can be reshaped to avoid obstacles by moving purely in the space of these additional degrees of freedom (null space). Although state-of-the-art robots allow surgeons to hand-guide endoscopic tools, reshaping the robot in null space is not intuitive for surgeons. Here we propose a learned task space control that allows surgeons to intuitively teach preferred robot configurations (shapes) that avoid obstacles using a VR-based planner in simulation. Later during surgery, surgeons control both the endoscopic tool and robot configuration (shape) with one hand. In a user study, we found that learned task space control outperformed state-of-the-art naive task space control in all the measured performance metrics (time, effort, and user-perceived effort). Our solution allowed users to intuitively interact with robots in VR and reshape robots while moving tools in medical and industrial applications.
Hollow fiber-based strain sensors with desirable modulus and sensitivity at effective deformation for dexterous electroelastomer cylindrical actuator
The electroelastomer cylindrical actuators, a typical representation of soft actuators, have recently aroused increasing interest owing to their advantages in flexibility, deformability, and spatial utilization rate. Proprioception is crucial for controlling and monitoring the shape and position of these actuators. However, most existing flexible sensors have a modulus mismatch with the actuation unit, hindering the free movement of these actuators. Herein, a low-modulus strain sensor based on laser-induced cellular graphitic flakes (CGF) onto the surface of hollow TPU fibers (HTF) is present. Through the electrostatic self-assembly technology, the flexible sensor features a unique hybrid sensing unit including soft HTF as substrate and rigid CGF as conductive path. As a result, the sensor simultaneously possesses desirable modulus (~0.155 MPa), a gauge factor of 220.3 (25% < ε < 50%), fast response/recovery behaviors (31/62 ms), and a low detection limit (0.1% strain). Integrating the sensor onto the electroelastomer cylindrical actuators enables precise measurement of deformation modes, directions, and quantity. As proof-of-concept demonstrations, a prototype soft robot with high-precision perception is successfully designed, achieving real-time detection of its deformations during the crawling process. Thus, the proposed scheme sheds new light on the development of intelligent soft robots.
Deep Bayesian active learning using in-memory computing hardware
Labeling data is a time-consuming, labor-intensive and costly procedure for many artificial intelligence tasks. Deep Bayesian active learning (DBAL) boosts labeling efficiency exponentially, substantially reducing costs. However, DBAL demands high-bandwidth data transfer and probabilistic computing, posing great challenges for conventional deterministic hardware. Here we propose a memristor stochastic gradient Langevin dynamics in situ learning method that uses the stochastic of memristor modulation to learn efficiency, enabling DBAL within the computation-in-memory (CIM) framework. To prove the feasibility and effectiveness of the proposed method, we implemented in-memory DBAL on a memristor-based stochastic CIM system and successfully demonstrated a robot’s skill learning task. The inherent stochastic characteristics of memristors allow a four-layer memristor Bayesian deep neural network to efficiently identify and learn from uncertain samples. Compared with cutting-edge conventional complementary metal-oxide-semiconductor-based hardware implementation, the stochastic CIM system achieves a remarkable 44% boost in speed and could conserve 153 times more energy.
Sensing-actuating integrated asymmetric multilayer hydrogel muscle for soft robotics
Achieving autonomously responding to external stimuli and providing real-time feedback on their motion state are key challenges in soft robotics. Herein, we propose an asymmetric three-layer hydrogel muscle with integrated sensing and actuating performances. The actuating layer, made of p(NIPAm-HEMA), features an open pore structure, enabling it to achieve 58% volume shrinkage in just 8 s. The customizable heater allows for efficient programmable deformation of the actuating layer. A strain-responsive hydrogel layer, with a linear response of up to 50% strain, is designed to sense the deformation process. Leveraging these actuating and sensing capabilities, we develop an integrated hydrogel muscle that can recognize lifted objects with various weights or grasped objects of different sizes. Furthermore, we demonstrate a self-crawling robot to showcase the application potential of the hydrogel muscle for soft robots working in aquatic environments. This robot, featuring a modular distributed sensing and actuating layer, can autonomously move forward under closed-loop control based on self-detected resistance signals. The strategy of modular distributed stimuli-responsive sensing and actuating materials offers unprecedented capabilities for creating smart and multifunctional soft robotics.
Responses