An Improved Alpha Beta Filter Using a Deep Extreme Learning Machine

This paper introduces new learning to the prediction model enhance algorithms’ performance in dynamic circumstances. We have proposed a novel technique based on alpha-beta filter and deep extreme machine (DELM) algorithm named as filter. The method has two main components, namely unit unit. used unit, uses DELM. problem with conventional is that values are generally selected via trial-and-error technique. Once chosen for specific problem, they remain fixed entire data. It been observed different same give results. Hence it essential tune according their historical behavior certain values. Therefore, method, we addressed this added module $\alpha $ - notation="LaTeX">$\beta improve filter’s performance. DELM algorithm’s dynamically changing conditions. measured using indoor environmental of temperature humidity. relative improvement model’s accuracy was 7.72% 16.47% RMSE MSE metrics. results show outperforms terms result compared