98%
921
2 minutes
20
Reducing energy consumption of wheeled robots in urban inspection and unstructured environments is a pressing challenge. This study proposes a human-like trajectory planning method based on deep learning to address energy inefficiency. A convolutional neural network (CNN) with multi-dimensional attention extracts spatial features from driving scenes and radar maps of hazardous areas. Temporal dependencies are captured using an improved long short-term memory (LSTM), where state information is added to the gate update module. Power, speed, and angular velocity are incorporated as constraints to enhance trajectory mapping accuracy. Experimental results show that, compared with traditional and state-of-the-art methods, the proposed approach significantly reduces cumulative power consumption and improves accuracy in predicting future trajectories. The model effectively learns human manipulation behaviors and demonstrates superior energy-saving performance in complex driving scenarios.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12397915 | PMC |
http://dx.doi.org/10.1016/j.isci.2025.113296 | DOI Listing |
Dev Sci
November 2025
Department of Psychology, University of Hong Kong, Pokfulam, Hong Kong Special Administrative Region.
Gender nonconforming (GN) children are at higher risk of experiencing bullying and social exclusion than are gender conforming (GC) children. Nonetheless, very little is known about the socio-cognitive mechanisms underlying children's bias against GN peers. The present study was the first to examine children's dehumanization of GN peers (developmental trajectory, form, and link to bullying).
View Article and Find Full Text PDFiScience
September 2025
College of Instrument Science and Electrical Engineering, Jilin University, Jilin, China.
Reducing energy consumption of wheeled robots in urban inspection and unstructured environments is a pressing challenge. This study proposes a human-like trajectory planning method based on deep learning to address energy inefficiency. A convolutional neural network (CNN) with multi-dimensional attention extracts spatial features from driving scenes and radar maps of hazardous areas.
View Article and Find Full Text PDFIEEE Int Conf Rehabil Robot
May 2025
In rehabilitation robotics, adapting exoskeleton behaviour to align with therapists' approach -where they intuitively vary their arm stiffness to provide tailored assistance to patients- is key for effective treatment and technology acceptance. Advanced robotic platforms have multiple setting parameters, but these often fail to precisely reflect therapists' desires. This work presents a novel framework to quantify and replicate therapists' stiffness when guiding arm movements, enabling its transfer to exoskeleton control.
View Article and Find Full Text PDFAffective touch is crucial for human development, social bonding, and emotional support, while physical interaction plays a key role in rehabilitation, aiding motor recovery and well-being. Although AI has advanced verbal and non-verbal communication, replicating human-like physical interaction remains a challenge. This paper explores whether robots can convey emotional touch.
View Article and Find Full Text PDFThis paper addresses the challenge of designing human-like reference trajectories for exoskeleton-aided rehabilitation, with a focus on mimicking human joint coordination while addressing clinical requirements. Redundant kinematic chains in human biomechanics pose challenges to trajectory planning: state-of-the-art algorithms often do not explicitly address the problem of replicating natural movements nor do they provide a suitable performance over a wide range of human motions. To address this challenge, this paper proposes a geodesics-based computational method that incorporates joint-level constraints, in addition to energy and level of comfort criteria to solve the problem of redundancy and better emulate human movements.
View Article and Find Full Text PDF