Biology subjects

Laschowski, B.

Publications and source records attributed to Laschowski, B..

21 records · Page 2Linked to original sources

bioRxiv2021

Computer Vision and Deep Learning for Environment-Adaptive Control of Robotic Lower-Limb Exoskeletons

Robotic exoskeletons require human control and decision making to switch between different locomotion modes, which can be inconvenient and cognitively demanding. To support the development of automated locomotion mode recognition systems (i.e., high-level controllers), we designed an environment recognition system using computer vision and deep learning. We collected over 5.6 million images of indoor and outdoor real-world walking environments using a wearable camera system, of which ~923,000 images were annotated using a 12-class hierarchical labelling architecture (called the ExoNet database). We then trained and tested the EfficientNetB0 convolutional neural network, designed for efficiency using neural architecture search, to predict the different walking environments. Our environment recognition system achieved ~73% image classification accuracy. While these preliminary results benchmark Efficient-NetB0 on the ExoNet database, further research is needed to compare different image classification algorithms to develop an accurate and real-time environment-adaptive locomotion mode recognition system for robotic exoskeleton control.

bioengineering↗

bioRxiv2020

ExoNet Database: Wearable Camera Images of Human Locomotion Environments

Advances in computer vision and artificial intelligence are allowing researchers to develop environment recognition systems for powered lower-limb exoskeletons and prostheses. However, small-scale and private training datasets have impeded the widespread development and dissemination of image classification algorithms for classifying human walking environments. To address these limitations, we developed ExoNet - the first open-source, large-scale hierarchical database of high-resolution wearable camera images of human locomotion environments. Unparalleled in scale and diversity, ExoNet contains over 5.6 million RGB images of different indoor and outdoor real-world walking environments, which were collected using a lightweight wearable camera system throughout the summer, fall, and winter seasons. Approximately 923,000 images in ExoNet were human-annotated using a 12-class hierarchical labelling architecture. Available publicly through IEEE DataPort, ExoNet offers an unprecedented communal platform to train, develop, and compare next-generation image classification algorithms for human locomotion environment recognition. Besides the control of powered lower-limb exoskeletons and prostheses, applications of ExoNet could extend to humanoids and autonomous legged robots.

bioengineering↗

bioRxiv2019

Biomechanical Modelling of Sitting Movements for Designing Robotic Lower-Limb Prostheses and Exoskeletons with Energy Regeneration

Although regenerative actuators can extend the operating durations of robotic lower-limb exoskeletons and prostheses, these energy-efficient powertrains have been exclusively designed and evaluated for continuous level-ground walking. ObjectiveHere we analyzed the lower-limb joint mechanical power during stand-to-sit movements using inverse dynamic simulations to estimate the biomechanical energy available for electrical regeneration. MethodsNine subjects performed 20 sitting and standing movements while lower-limb kinematics and ground reaction forces were measured. Subject-specific body segment parameters were estimated using parameter identification, whereby differences in ground reaction forces and moments between the experimental measurements and inverse dynamic simulations were minimized. Joint mechanical power was calculated from net joint torques and rotational velocities and numerically integrated over time to determine joint biomechanical energy. ResultsThe hip produced the largest peak negative mechanical power (1.8 {+/-} 0.5 W/kg), followed by the knee (0.8 {+/-} 0.3 W/kg) and ankle (0.2 {+/-} 0.1 W/kg). Negative mechanical work from the hip, knee, and ankle joints per stand-to-sit movement were 0.35 {+/-} 0.06 J/kg, 0.15 {+/-} 0.08 J/kg, and 0.02 {+/-} 0.01 J/kg, respectively. Conclusion and SignificanceAssuming an 80-kg person and previously published regenerative actuator efficiencies (i.e., maximum 63%), robotic lower-limb exoskeletons and prostheses could theoretically regenerate ~26 Joules of total electrical energy while sitting down, compared to ~19 Joules per walking stride. Given that these regeneration performance calculations are based on healthy young adults, future research should include seniors and/or rehabilitation patients to better estimate the biomechanical energy available for electrical regeneration among individuals with mobility impairments.

bioengineering↗