Advancements in Humanoid Robots: A Comprehensive Review and Future Prospects

IEEE/CAA Journal of Automatica Sinica, Год журнала: 2024, Номер 11(2), С. 301 - 328

Опубликована: Янв. 29, 2024

This paper provides a comprehensive review of the current status, advancements, and future prospects humanoid robots, highlighting their significance in driving evolution next-generation industries. By analyzing various research endeavors key technologies, encompassing ontology structure, control decision-making, perception interaction, holistic overview state robot is presented. Furthermore, emerging challenges field are identified, emphasizing necessity for deeper understanding biological motion mechanisms, improved structural design, enhanced material applications, advanced drive methods, efficient energy utilization. The integration bionics, brain-inspired intelligence, mechanics, underscored as promising direction development robotic systems. serves an invaluable resource, offering insightful guidance to researchers field, while contributing ongoing potential robots across diverse domains.

Язык: Английский

---

Brain-inspired multimodal hybrid neural network for robot place recognition

Science Robotics, Год журнала: 2023, Номер 8(78)

Опубликована: Май 10, 2023

Place recognition is an essential spatial intelligence capability for robots to understand and navigate the world. However, recognizing places in natural environments remains a challenging task because of resource limitations changing environments. In contrast, humans animals can robustly efficiently recognize hundreds thousands different conditions. Here, we report brain-inspired general place system, dubbed NeuroGPR, that enables by mimicking neural mechanism multimodal sensing, encoding, computing through continuum space time. Our system consists hybrid network (MHNN) encodes integrates cues from both conventional neuromorphic sensors. Specifically, encode sensory cues, built various networks view cells, head direction time cells. To integrate these designed multiscale liquid state machine process fuse information effectively asynchronously using diverse neuronal dynamics bioinspired inhibitory circuits. We deployed MHNN on Tianjic, chip, integrated it into quadruped robot. results show NeuroGPR achieves better performance compared with existing biologically inspired approaches, exhibiting robustness environmental uncertainty, including perceptual aliasing, motion blur, light, or weather changes. Running as overall multi–neural workload Tianjic showcases its advantages 10.5 times lower latency 43.6% power consumption than commonly used mobile robot processor Jetson Xavier NX.

Язык: Английский

---

Design and performance analysis of different cambered wings for flapping-wing aerial vehicles based on wind tunnel test

Robotic Intelligence and Automation, Год журнала: 2024, Номер 44(5), С. 698 - 712

Опубликована: Июль 11, 2024

Purpose Bionic flapping-wing aerial vehicles (FWAVs) mimic natural flyers to generate the lift and thrust, such as birds, bats insects. As an important component of FWAVs, flapping wings are crucial for flight performance. The aim this paper is study effects different on aerodynamic Design/methodology/approach Inspired by structure authors design four cambered analyze effect airfoils FWAVs motor-driven mechanism wings, realize control frequency. Combined with wind tunnel equipment, build force test platform static dynamic performance under state variables frequency, speed inclined angle. Findings results show that wing a camber 20 mm best. Compared flat wing, average can be improved 59.5%. Originality/value Different from traditional given in paper. influence analyzed optimal obtained.

Язык: Английский

---

Large language models for robotics: Opportunities, challenges, and perspectives

Journal of Automation and Intelligence, Год журнала: 2024, Номер unknown

Опубликована: Дек. 1, 2024

Язык: Английский

---

Comprehensive Nonlinear Control Strategy for VTOL-UAVs With Windowed Output Constraints

IEEE Transactions on Control Systems Technology, Год журнала: 2023, Номер 31(6), С. 2673 - 2684

Опубликована: Июль 31, 2023

In this article, we develop a methodology for trajectory tracking control of vertical takeoff and landing (VTOL) unmanned aerial vehicle (UAV) in the presence unknown mass time-varying external disturbances. A time-dependent shift function is introduced, which, alongside set barrier vector functions, enables finite period constrained outputs, namely, position velocity errors. The proposed controller also features two stable adaption laws to estimate following: 1) UAV 2) parameter relating bound total closed-loop system error shown be uniformly ultimately bounded (UUB). This strategy, which requires only first derivative reference an implicit saturation thrust force, especially apt handle three stages flight: takeoff, tracking, landing. Simulation results are presented validate solution. Experimental included showcase performance our real-world application using quadrotor vehicle.

Язык: Английский

---

A review on energy efficiency in autonomous mobile robots

Robotic Intelligence and Automation, Год журнала: 2023, Номер 43(6), С. 648 - 668

Опубликована: Сен. 14, 2023

Purpose This paper aims to provide a comprehensive analysis of the state art in energy efficiency for autonomous mobile robots (AMRs), focusing on sources, consumption models, energy-efficient locomotion, hardware consumption, optimization path planning and scheduling methods, suggest future research directions. Design/methodology/approach The systematic literature review (SLR) identified 244 papers analysis. Research articles published from 2010 onwards were searched databases including Google Scholar, ScienceDirect Scopus using keywords search criteria related power management various robotic systems. Findings highlights following key findings: batteries are primary source AMRs, with advances battery systems enhancing efficiency; hybrid models offer superior accuracy robustness; locomotion contributes over 50% robot’s total emphasizing need optimized control methods; factors such as center mass impact AMR consumption; algorithms methods essential optimization, algorithm choice depending specific requirements constraints. limitations/implications concentrates wheeled robots, excluding walking ones. Future work should improve explore examine artificial intelligence/machine learning roles assess trade-offs. Originality/value provides highlighting findings SLR suggests directions further advancements this field.

Язык: Английский

---

Design and Quantitative Assessment of Teleoperation-Based Human–Robot Collaboration Method for Robot-Assisted Sonography

IEEE Transactions on Automation Science and Engineering, Год журнала: 2024, Номер unknown, С. 1 - 11

Опубликована: Янв. 1, 2024

Tele-echography has emerged as a promising and effective solution, leveraging the expertise of sonographers autonomy robots to perform ultrasound scanning for patients residing in remote areas, without need in-person visits by sonographer. Designing natural human-robot interfaces tele-echography remains challenging, with patient safety being critical concern. In this article, we develop teleoperation system robot-assisted sonography two different interfaces, haptic device-based interface low-cost 3D Mouse-based interface, which can achieve continuous intuitive telemanipulation leader device small workspace. To compliant interaction patients, design impedance controllers Cartesian space track desired position orientation these interfaces. We also propose comprehensive evaluation metrics sonography, including subjective objective evaluation, evaluate control performance. ergonomic performance based on estimated muscle fatigue acquired image quality. conduct user studies NASA Task Load Index The tracking quantitative comparison are conducted Franka Emika Panda robot. results findings provide guidance collaboration implementation sonography. Note Practitioners —Robot-assisted demonstrated efficacy medical diagnosis during clinical trials. However, deploying fully autonomous challenging due various constraints practice, such safety, dynamic tasks, environmental uncertainties. Semi-autonomous or teleoperation-based robot represents approach practical deployment. Previous work produced expensive but lacks guide selection. present typical implement system. metric assessing their Our show that outperforms Mouse, operators' feedback requires more learning time effort training stage. Furthermore, developed offers solution shared skill transfer. valuable designing implementing practice.

Язык: Английский

---

USTButterfly: A Servo-Driven Biomimetic Robotic Butterfly

IEEE Transactions on Industrial Electronics, Год журнала: 2023, Номер 71(2), С. 1758 - 1767

Опубликована: Март 27, 2023

Different from birds, bats, flies, and other flying creatures that generate propulsions by flapping a pair of wings, butterflies with two pairs wings have their peculiar flight aerodynamics. In this article, servo-driven biomimetic robotic butterfly, named USTButterfly, is designed to study the mechanism biological butterflies. First, instead using single actuator drive servos are introduced independently left right thereby realizing tailless control butterfly. Next, bionic airfoil inspired wing structure glasswing Wing geometry analysis indicates USTButterfly well matches morphological characteristics Finally, multicamera motion capture system used measure results demonstrate despite larger Reynolds number, shares similar coupled wing–body interaction in climbing flight. opens up new way specializations butterflies, offers potential alternative paradigm for existing unmanned aerial vehicles.

Язык: Английский

---

Adaptive Internal Model Control for a Flexible Wing With Unsteady Aerodynamic Loads

IEEE Transactions on Cybernetics, Год журнала: 2023, Номер 54(1), С. 546 - 557

Опубликована: Июнь 5, 2023

This article proposes adaptive internal model controls for the collocated output regulation of a flexible wing, where distributed disturbances, boundary and references are from an exactly unknown exosystem. Observer-based tracking error feedback first designed to address robust in case known exosystem matrix. If has matrix, observer is further proposed with system converging zero exponentially. Then, we can obtain observer-based by combining observers controls, which able regulate errors toward disturbances references. The corresponding closed-loop proved be internally asymptotically stable. A simulation example provided control wing system.

Язык: Английский

---

A novel human-robot skill transfer method for contact-rich manipulation task

Robotic Intelligence and Automation, Год журнала: 2023, Номер 43(3), С. 327 - 337

Опубликована: Июнь 6, 2023

Purpose The purpose of this paper is to enhance the robot’s ability complete multi-step contact tasks in unknown or dynamic environments, as well generalization same task different environments. Design/methodology/approach This proposes a framework that combines learning from demonstration (LfD), behavior tree (BT) and broad system (BLS). First, original motion primitive modified have better for representing primitives. Then, BT based on constructed, which will select appropriate primitives according environment state robot ontology state, then BLS generate specific parameters state. weights can also be optimized after each successful execution. Findings authors carried out cleaning desktop assembling shaft hole Baxter Elite robots, respectively, both were successfully completed, proved effectiveness framework. Originality/value LfD, BLS. To best authors’ knowledge, no similar methods found other people’s work. Therefore, believe work original.

Язык: Английский

---