A 3D Computer Vision-Guided Robotic Companion for Non-Contact Human Assistance and Rehabilitation

Shen, Tao; Afsar, Rayhan; Zhang, He; Ye, Cang; Shen, Xiangrong

doi:10.1007/s10846-020-01258-1

Cited by 6 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Particularly, through the use of RGB-D cameras, usually built into SWs for other functionalities (e.g., gait and posture assessment (Palermo et al, 2021)). Shen et al (2020) used a RGB-D camera to perceive the intended walking direction from the torso kinematics. An hybrid Proportional-Integral-Derivative (PID) controller with integrated digital practical differentiator was then implemented to calculate the desired wheel rotation angles, being able to track the subject in forward and turning movements.…”

Section: Human Motion Decoding In Smart Walkersmentioning

confidence: 99%

“…An hybrid Proportional-Integral-Derivative (PID) controller with integrated digital practical differentiator was then implemented to calculate the desired wheel rotation angles, being able to track the subject in forward and turning movements. However, the use of upper body information entailed a disadvantage: the upper body swaying (Shen et al, 2020). Chalvatzaki et al (2019) complements RGB-D data with 2D laser data to perform real-time gait analysis, as well as to track the user's Centre-of-Mass position and velocity.…”

Section: Human Motion Decoding In Smart Walkersmentioning

confidence: 99%

“…The i-Walk system (Chalvatzaki et al, 2020) points out the potential of using RGB-D cameras, in particular coupled with DL models, to attain an effective activity and gesture recognition, but not specifically to drive the walker. Although promising, these vision-based motion decoding approaches (Shen et al, 2020;Chalvatzaki et al, 2019Chalvatzaki et al, , 2020 require additional computational tasks to estimate human poses or parameters from the acquired RGB videos. This can lead to error propagation, when feeding these inputs to a following motion decoding algorithm.…”

Section: Human Motion Decoding In Smart Walkersmentioning

confidence: 99%

See 2 more Smart Citations

Deep learning-based approaches for human motion decoding in smart walkers for rehabilitation

Gonçalves¹,

Lopes²,

Moccia³

et al. 2023

Preprint

View full text Add to dashboard Cite

Gait disabilities are among the most frequent impairments worldwide. Their treatment increasingly relies on rehabilitation therapies, in which smart walkers are being introduced to empower the user's recovery state and autonomy, while reducing the clinicians effort. For that, these should be able to decode human motion and needs, as early as possible. Current walkers decode motion intention using information gathered from wearable or embedded sensors, namely inertial units, force sensors, hall sensors, and lasers, whose main limitations imply an expensive solution or hinder the perception of human movement. Smart walkers commonly lack an advanced and seamless human-robot interaction, which intuitively and promptly understands human motions. A contactless approach is proposed in this work, addressing human motion decoding as an early action recognition/detection problematic, using RGB-D cameras. We studied different deep learning-based algorithms, organised in three different approaches, to process lower body RGB-D video sequences, recorded from an embedded camera of a smart walker, and classify them into 4 classes (stop, walk, turn right/left). A custom dataset involving 15 healthy participants walking with the walker device was acquired and prepared, resulting in 28800 balanced RGB-D frames, to train and evaluate the deep learning networks. The best results were attained by a convolutional neural network with a channel-wise attention mechanism, reaching accuracy values of 99.61% and above 93%, for offline early detection/recognition and trial simulations, respectively. Following the hypothesis that human lower body features encode prominent information, fostering a more robust prediction towards real-time applications, the algorithm focus was also quantitatively evaluated using Dice metric, leading to values slightly higher than 30%. Promising results were attained for early action detection as a human motion decoding strategy, with enhancements in the focus of the proposed architectures.

show abstract

Section: Human Motion Decoding In Smart Walkersmentioning

confidence: 99%

Section: Human Motion Decoding In Smart Walkersmentioning

confidence: 99%

Section: Human Motion Decoding In Smart Walkersmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning-based approaches for human motion decoding in smart walkers for rehabilitation

Gonçalves¹,

Lopes²,

Moccia³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The device is evaluated through laboratory tests and the results demonstrate its ability to accurately detect pathogens and effectively release disinfectants. [31] aim at helping the mobility-challenged individuals with a novel robotic companion, which is a walker-type mobile robot capable of accompanying the human user and keeping user at the center for protection and possible power assistance. The authors propose a robot that can assist mobility-challenged individuals with navigation, object recognition, and power assistance.…”

Section: Introductionmentioning

confidence: 99%

A Low-Cost Wearable Autonomous System for the Protection of Bicycle Users

Mejía¹,

Gómez²,

Sarmiento³

2023

IJACSA

View full text Add to dashboard Cite

A bicycle is a form of transport that not only positively impacts the health of users, and the general population by reducing pollution levels but also constitutes an accessible and affordable means of transport for developing societies. However, when coexisting with other forms of transport, the accident rate is elevated, and the risk is high. Among the factors contributing to accidents involving bicycles are collisions with motor vehicles. These accidents can occur when a motor vehicle maneuvers and does not see the bicycle or when a motorist drives distracted. These types of accidents can be avoided if cyclists and motorists are aware of the environment and respect traffic laws and safety regulations. This research aims to develop a low-cost autonomous electronic system that provides extra protection to bicycle users, particularly by making them visible to other road users on cloudy days or at night. The system uses a 32-bit processor with brightness and acceleration sensors that trigger visual alerts to both the bicycle user and possible nearby vehicles. It also monitors and logs the signals on a server for route evaluation. The laboratory successfully evaluated the prototype, demonstrating its autonomy and performance. The test results obtained demonstrate the system's capacity to provide extra protection, in addition to its robustness and accuracy.

show abstract

Deep learning-based approaches for human motion decoding in smart walkers for rehabilitation

Gonçalves

Lopes

Moccia

et al. 2023

Expert Systems with Applications

View full text Add to dashboard Cite

A 3D Computer Vision-Guided Robotic Companion for Non-Contact Human Assistance and Rehabilitation

Cited by 6 publications

References 32 publications

Deep learning-based approaches for human motion decoding in smart walkers for rehabilitation

Deep learning-based approaches for human motion decoding in smart walkers for rehabilitation

A Low-Cost Wearable Autonomous System for the Protection of Bicycle Users

Deep learning-based approaches for human motion decoding in smart walkers for rehabilitation

Contact Info

Product

Resources

About