Humanoid robots offer a promising solution to the growing burden of care for older adults. However, existing evidence on their applications for general aging populations remains fragmented and lacks systematic synthesis. This scoping review aimed to examine the literature on humanoid robot-assisted support for health care in older adults and identify gaps in the literature to guide future research. The methodological framework by Arksey and O'Malley was used to conduct this scoping review. We conducted a comprehensive search in 8 databases, including IEEE Xplore Digital Library, CINAHL, Cochrane Library, EMBASE, PubMed, Scopus, Web of Science, and OpenGrey Repository, covering literature published up to April 30, 2025. The reference lists of key texts were examined, and citation chaining was conducted. Two independent reviewers examined all full articles for fitness with the eligibility criteria and extracted data elements. The study findings were then summarized, coded, and analyzed using the PAGER (Patterns, Advances, Gaps, Evidence for practice, and Research recommendations) framework. A total of 32,477 articles were retrieved, 59 of which were included in this review. The majority (49/59, 83%) were conducted in real-world settings. Methodologically, 34 studies (34/59, 58%) had small sample sizes (n≤25), with study designs comprising 26 quantitative (26/59, 44%), 22 mixed method (22/59, 37%), and 11 qualitative (11/59, 19%) approaches. Participant characteristics revealed female predominance (>50%) in 32 studies (32/59, 54%), while 27 studies (27/59, 46%) included participants with cognitive impairment. Through PAGER framework analysis, we identified 4 key patterns: (1) effects, perceptions, and experiences of humanoid robots; (2) preferences, expectations, and facilitators for humanoid robots; (3) implementation barriers and challenges; and (4) determinants of user experiences and outcomes. This scoping review demonstrates the promising yet methodologically constrained potential of humanoid robots in health care for older adults while highlighting key challenges in their practical implementation. Successful integration will require addressing technical limitations, user acceptance barriers, and systemic adoption challenges.
This study aims to enhance the precision of humanoid robots in imitating complex human "walking-grasping" coordinated movements. Addressing limitations in sample efficiency and reward function design in Generative Adversarial Imitation Learning (GAIL), we propose the Similarity Reward-Augmented Generative Adversarial Imitation Learning (SRA-GAIL) framework. The method integrates plantar thin-film resistive pressure sensors to measure the real-time pressure distribution at four key points on both feet, combined with roll/pitch angle data acquired from JY901S inertial measurement units (IMUs). A Lagrangian constraint optimization strategy is employed to achieve gait stability control based on the zero moment point (ZMP). Simultaneously, a visual similarity evaluation module is established using human demonstration trajectories captured by a Logitech C920E camera, augmented by grip force feedback from flexible thin-film pressure sensors on the hands. This enables the design of a multimodal sensor-fused similarity reward function. By incorporating Lagrangian constraint optimization and a maximum entropy reinforcement learning framework, Similarity Reward-Augmented Generative Adversarial Imitation Learning synchronously optimizes gait stability control-guided by zero moment point (ZMP) and roll/pitch data-and vision-based trajectory similarity evaluation. These components address motion stability constraints and trajectory similarity metrics, respectively, generating biomechanically plausible gait strategies. A spatiotemporal attention mechanism parses human motion trajectory features to drive the end-effector for high-precision trajectory tracking. To validate the proposed method, an imitation learning experimental system was constructed on a physical XIAOLI humanoid robot platform, integrating inertial measurement units (IMUs), plantar pressure sensors, and a vision system. Quantitative evaluations were conducted across multiple dimensions, including robot platform analysis, walking stability, object grasping success rates, and end-effector trajectory similarity. The results demonstrate that, compared to Generative Adversarial Imitation Learning (GAIL) and behavioral cloning, Similarity Reward-Augmented Generative Adversarial Imitation Learning achieves a stable object grasping success rate of 93.7% in complex environments, with a 23.8% improvement in sample efficiency. The method maintains a 96.5% compliance rate for zero moment point (ZMP) trajectories within the support polygon, significantly outperforming baseline approaches. This effectively addresses the bottleneck in robot policies adapting to dynamic changes in real-world environments.
Realistic reproduction of human facial expressions is essential for realistic interactions between humans and humanoid robots. This work presents a data-driven framework for transferring human facial expressions to a humanoid robot and a virtual avatar, aiming to enhance emotional expressiveness and assess its applicability in psychiatric training scenarios. The proposed approach enables cross-domain facial expression mapping while accounting for mechanical constraints of robotic actuation. A user study (n = 40) evaluated emotion recognition across three stimulus categories: human faces (H), unconstrained virtual avatars (A) and humanoid robots with limited facial actuation (R). Participants identified emotions from static images and from dynamic expression sequences, presented with and without speech. Perceived realism and uncanny valley effects were assessed using an eight-item questionnaire rated on a 7-point Likert scale. Results indicate that human-to-robot facial expression transfer is feasible but constrained by mechanical expressivity. Highly expressive emotions such as surprise (H: 87.5%; A: 57.5%; R: 65%) and fear (H: 45%; A: 27.5%; R: 57.5%) achieved moderate recognition rates, whereas subtle emotions such as anger (H: 65%; A: 40%; R: 12.5%) and disgust (H: 60%; A: 10%; R: 22.5%) were poorly recognized on the robot. Dynamic expressions combined with speech significantly improved recognition. These findings demonstrate the feasibility of transferring human facial expressions to humanoid robots while highlighting current limitations of robotic facial actuation. The proposed framework provides a promising basis for emotionally realistic patient simulation and training applications in mental healthcare.
Mechanical stimulation is essential in tissue engineering and regenerative medicine for proper tissue maturation. However, conventional uniaxial platforms fail to reproduce the multiaxial loading experienced in vivo. In this study, we present a humanoid robotic bioreactor capable of delivering human-like shoulder motions to engineered tendon constructs, enabling controlled multiaxial stimulation with real-time strain monitoring. Human mesenchymal stem cells were cultured on decellularized tendon scaffolds and subjected to adduction-abduction loading at peak strains of approximately 3.5% and 9.5% under external forces of 25 and 50 N, respectively. Strain levels were directly quantified in situ using a flexible sensor integrated within the bioreactor. The transparent bioreactor membrane allowed noninvasive observation while simultaneously applying mechanical stimulation over 14 d, with continuous assessment of cellular morphology without fixation. Compared with static and traditional uniaxial controls, the robot motions enhance cell alignment and activation of mechanotransduction pathways while inducing notable gene and protein expression changes, particularly within the PI3K-Akt signaling pathway. Although dynamic loading resulted in a moderate reduction in cell viability, the transcriptional profile was consistent with mechanically driven phenotypic adaptation toward tenogenic-related programs rather than dominant signatures of acute cytotoxic damage. These findings demonstrate that replicating human-like multiaxial mechanics in vitro fundamentally alters cellular mechanosensing and may provide a mechanobiological foundation for the future development of more physiologically relevant tendon grafts.
Socially assistive humanoid robots have emerged as promising tools for enhancing mood, engagement, and social connection in clinical and adult day care environments. This study investigated the acceptance and emotional impact of a humanoid robot deployed over an eight-week period in adult day care facilities, with the goal of reducing depressive affect and improving positive mood through activities such as trivia, karaoke, dancing, and humor. Sixty-five residents participated in the intervention. Affective and behavioral engagement were measured using the Robot Engagement Rating Scale (RERS), supplemented by qualitative field observations. The study assessed changes in engagement, social behaviors (e.g., eye contact), and participants' desire to continue interaction with the robot. RERS scores showed a significant increase in engagement, rising from 3.09 ± 0.27 at baseline to 4.61 ± 0.12 post-interaction, t(14) = 5.15, p < .001. Eye contact with the robot increased from 66.7% to 94.4% (p = .041), and the proportion of participants requesting to extend their interaction grew from 63.6% to 83.3%. Qualitative observations described laughter, verbal reciprocity, and sustained attention throughout the sessions, confirming positive emotional engagement. Conclusion / Discussion: Findings demonstrate that humanoid robots are highly acceptable and effective in eliciting positive emotional engagement among older adults in day care settings. These results underscore their therapeutic potential as complementary, socially enriching tools for promoting well-being and mood stabilization in long-term care and rehabilitation environments.
Humanoid upper-limb robots are an important direction in biomimetic robotics, and inverse kinematics is a key technique for achieving human-like coordinated operation. However, existing inverse kinematics methods for bimanual trajectory tracking often suffer from high computational complexity and limited synchronization performance. To address this, this paper proposes an error-adaptive competition-based inverse kinematics (EAC-IK) approach for bimanual trajectory tracking of humanoid upper-limb robots. First, a unified modeling framework for the absolute tracking errors and synchronization errors of the two arms is established, and the end-effector task constraints are reformulated into a low-dimensional representation, thereby reducing the computational complexity of the original high-dimensional task mapping. Second, to enhance the coordination capability of bimanual operations, an error-adaptive competition mechanism is developed to regulate the weighting coefficients of the two arms online according to their error states. In addition, a virtual second-order command shaper is introduced at the joint level to reconstruct joint trajectories and suppress oscillations induced by input noise and the error-adaptive competition mechanism. Simulation and experimental results on a hyper-redundant humanoid upper-limb robot demonstrate that, compared with the zeroing neural-network-based inverse kinematics method, the proposed method achieves lower tracking and synchronization errors, as well as higher computational efficiency. In the circular trajectory-tracking experiment, the left-arm position and orientation tracking errors decrease from 1.60×10-3m and 4.72×10-3rad to 0.70×10-3m and 0.95×10-3rad, respectively, while the synchronization error decreases from 1.96×10-3 to 1.30×10-3. In addition, the average algorithm runtime decreases from 0.82ms to 0.63ms.
暂无摘要（点击查看详情）
Humanoid robots can be seamlessly integrated into human-robot interaction scenarios due to their human-like appearances. Pneumatic artificial muscles (PAM) are promising actuators for such robots due to their similarity to biological muscles, but their limited contraction ratio constrains both appearance and motion range of the robot. This work presents a 7-degrees-of-freedom (DoF) pneumatic humanoid robotic arm that mimics the human arm in both appearance and movement capabilities. A hybrid actuation scheme, combining direct PAM actuation at shoulder joints and PAM actuation with Bowden cable transmission at the distal joints, is adopted to enable anthropomorphic scaling with a lightweight and compliant structure. To address the control challenges posed by the nonlinear dynamics of PAMs and Bowden cables, Pneumatic Bowden cable Optimized Soft Actor-Critic (PBO-SAC), a model-free reinforcement learning framework, is proposed to enable efficient on-hardware control policy learning for the robotic arm. PBO-SAC incorporates posture-perturbed decoupled training and local recurrent fusion networks to ensure safe and smooth policy learning. Simulation results verify improvements in PBO-SAC, while hardware experiments on trajectory tracking and teleoperated stacking tasks further demonstrate the multi-DoF coordination control performance.
At the 2025 International Conference of Intelligent Robots and Systems (IROS) in Zhejiang, China, leading researchers and industry experts debated whether humanoid robots will soon replace human workers. A summary of the points raised by the debaters for or against the claim is highlighted below.
Drug discovery research faces significant challenges, with around 90 % of candidate drugs failing in clinical trials. Traditional animal models do not always accurately reflect human physiology and pathology, leading to low predictability of clinical outcomes. In vitro human model using induced pluripotent stem cells (iPSCs), offers a potential solution, but long culture periods and variability among experiments have prompted a shift towards automation in human biomimetic research. We optimized the automated cell culture system "Maholo LabDroid" to establish a platform for the long-term culture of undifferentiated iPSCs and for conducting complex cell-based assays. The Maholo LabDroid, a humanoid robotic system with excellent reproducibility, allows for the addition of peripherals and devices to execute new protocols. To enhance the use of iPSCs, we addressed limitations in detecting real-time changes in cells, supplying sufficient high-quality cells and handling 3D cell models that reflect more physiologically relevant conditions. We integrated a robot arm-assisted imaging system for real-time observation, introduced Bluetooth-enabled electronic pipettes for flexible liquid handling, and expanded the types of culture containers used. These improvements enabled us to determine optimal culture conditions, obtain sufficient quantities of cells, and automatically generate organoids with physiological functions. Using the LabDroid Maholo, we built a platform technology for the provision of advanced cellular models. This platform has the potential to enhance drug discovery research and contribute to the development of cell therapy products.
Acceptance of robots depends on whether they support consumers' psychological needs, making demand-side insights essential.
暂无摘要（点击查看详情）
暂无摘要（点击查看详情）
暂无摘要（点击查看详情）
Target characteristics can influence motor performance in goal-directed tasks by affecting cognitive and affective processes. This study investigated whether motor performance in a ball-throwing task differed when the target was held by a person, compared with when the target was a physical mark. Eighteen healthy adults without baseball experience participated. Each participant threw 15 fastballs across three blocks, each with a different target condition. In the Mark condition, the target was a visible mark on the net. In the Humanoid condition, the same mark was placed on a humanoid board. In the Catcher condition, the mark was held by a person. Mean ball speed, absolute error of arrival positions, pupil diameter, and subjective ease of throwing were compared across conditions. Mean ball speed and absolute error were significantly lower in the Catcher condition than in the other two conditions. Conversely, pupil diameter and subjective ease were significantly higher in the Catcher condition. These results indicate that when the target is held by a person, participants prioritized accuracy over speed. Increased pupil diameter in the pre-movement state suggests heightened arousal in the Catcher condition, which was associated with improved accuracy. Subjective ease also increased in this condition. These findings suggest that an implicit social or safety-related goal of allowing the person to catch the ball, such as safety and harm-avoidance considerations for the catcher, led participants to focus more on precise throws, contributing to the understanding of how cognitive and affective processes influence motor control.
This paper proposes a learning-based contact force controller using deep neural networks (DNN) and a PI controller. Stable contact force control between the foot and the ground is essential for humanoid robots to maintain balance during bipedal walking. While admittance controllers have been extensively employed for contact force control in humanoid robots, their performance is limited by the high nonlinearity inherent in robot systems. To overcome these limitations, we propose a deep neural network (DNN)-based inverse model, which leverages input-output data that inherently capture system nonlinearities. The proposed learning-based contact force controller computes the target foot height based on the target force, measured force, and measured foot height, without relying on a dynamic model of the articulated robotic leg. Furthermore, a PI controller is integrated to mitigate steady-state errors. Experimental comparisons between the proposed controller and an admittance controller were conducted using an articulated robotic leg. Compared with an admittance controller, the proposed method reduced overshoot by 96% and settling time by 61% on average in step responses and decreased force-tracking RMSE by 66.3% on average across both step and sinusoidal experiments.
Visual perspective taking (VPT) is a crucial component of social cognition: when engaging in activities such as preparing meals together, moving a couch together, playing tennis, and discussing our surroundings, the ability to keep track of what others see is essential. In recent years, studies investigated the cognitive mechanisms of VPT not only of other humans but also of artificial agents, such as robots. Thus far, researchers have primarily investigated interference tasks with regard to artificial agents (i.e., how the perspective of an agent interferes with one's own perspective). In the present study, we investigated a task where VPT has a facilitatory effect (i.e., taking the perspective of an agent results in a performance boost) across four experiments, in which we varied the type of the avatar: a human (Exp. 1), a humanoid robot with a human-like head (Exp. 2), a humanoid robot with a camera-like head (Exp. 3), and a camera (Exp. 4). We found significant VPT effects for the human and robot avatar but failed to do so for the robot with a camera-like head and the camera. These findings suggest that participants adopted the perspective of avatars that are either human or resemble a human-like appearance. However, this is not the case for avatars which deviate from a human-appearance or which imply a social presence.
In immersive VR applications, avatars shape users' identities and social presence. Personalization requires generating realistic outfits for avatars, often specified through a single reference image, while supporting seamless editing, adaptation to diverse avatars, and efficient rendering. Achieving these goals is challenging because VR avatars, even humanoid ones, exhibit substantial variations in body shapes and topologies. This inherent diversity makes it difficult to collect sufficient paired data, impeding the evolution of generalizable end-to-end image-to-garment models. To this end, we propose Tailor, a two-stage framework that dresses 3D humanoid avatars from a single reference image while preserving the wearing effects observed in the image. In the first stage, Tailor leverages a structured garment representation based on sewing patterns, enabling the network to predict garments in a low-dimensional, interpretable, and topology-independent space. In the second stage, Tailor performs instance-specific optimization to adapt the predicted sewing pattern to the avatar, ensuring consistent wearing effects across varying avatars. Furthermore, this framework also enables seamless garment editing, on-the-fly adaptation, and real-time rendering, making it particularly suitable for large-scale VR environments. Extensive experiments demonstrate that Tailor achieves results comparable to professional manual designs and produces garments that are both visually appealing and better aligned with reference styles than those generated by naive pattern-scaling baselines, as validated through human perceptual studies.
The 2025 novel Mechanize My Hands for War features humanoid robots for agriculture.