Visual based sensor cart follower for wheelchair by using microcontroller

Ahmad, Muhammad Shakil; Alhady, S. S. N.; Kaharuddin, S.; Othman, W. A. F. W.

doi:10.1109/iccsce.2015.7482170

Cited by 7 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CMUcam used as a tracking sensor contributed to a successful autonomous mobile robot. The taken value of camera's field of view (FOV) with several details from other sensors gathered as well to establish the robust tracking system [10][11][12][13].…”

Section: Mobile Robotmentioning

confidence: 99%

Embedded Artificial Neural Network FPGA Controlled Cart

Ahmad¹,

Alhady²,

Oon³

et al. 2019

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

An artificial neural network (ANN) computing system can be significantly influenced by its implementation type. The software implemented ANN can produce high accuracy output with slow computation time performance compared to hardware implemented ANN which runs at a faster computation time but with low accuracy. Normally, software implementation reduces the proficiency and efficiency of the model. Robot performance plays an important role as it needs fast response to process information that is applied with ANN. As a consequence, the proposed research focuses on comparison between hardware and software implementation multilayer perceptron (MLP) for cart follower in Field Programmable Logic Array (FPGA). Both of the software and hardware models produced the same precision where the output distance at angles-10°, 0° and 10° shows same percentage error. Besides that, both of the models have similar root mean square error (RMSE) which are 0.469, 0.479 and 0.267 at-10°, 0° and 10° respectively. The processing time of MLP model implemented in hardware and software are at 1.91μs and 78.06μs respectively. Thus, it can be concluded that hardware implementation is better than software implementation.

show abstract

Section: Mobile Robotmentioning

confidence: 99%