Journal DOI 10.68956/journal.human-performance-futures

Creative Commons source reading · Explainable decision support

Artificial Intelligence-Driven Neural Computing for Navigation Decision Classification in Autonomous Vehicles

Mohammed Baihan

Originally published in Biomimetics (Basel) · 2026-10-03

From the article

Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.

Abstract

Autonomous vehicles (AVs) require robust decision-making capabilities to navigate dynamic environments securely and effectively. Furthermore, advanced sensors and computational models are utilized to perceive the environment and predict the behavior of other road users. The main challenges include dealing with unpredictable factors such as weather fluctuations, traffic patterns, and road obstacles. While these technologies provide contextual awareness, existing techniques often face difficulties in handling complex sensor data under rapidly changing conditions and selecting the most useful features for navigation decision classification. To overcome these challenges, there is a need for reliable and intelligent architectures for navigation decision classification systems to support AVs by utilizing AI-driven machine learning (ML) and…

4. Experimental Findings and Discussions

The empirical evaluation of the IBTAF-NDCAV method was conducted using the Autonomous Driving Sensor-Kinematics (ADSK) Dataset [26] and the ADAS-EV Dataset [27]. The dataset is a real-time dataset developed to help research and expansion in AV risk prediction and decision-making. This dataset can incorporate simulated sensor data (radar, LiDAR, and camera) with environmental contexts, vehicle kinematics, and navigation-relevant parameters. The dataset is implemented for ML method training and assessment, thus providing balanced traffic conditions, obstacle encounters, and environmental conditions. This dataset is especially valuable for evaluating navigation decision classification under dynamic settings. It is also useful for real-world navigation decision classification, risk prediction and collision systems, and multi-sensor data…

5. Conclusions

This article presented the IBTAF-NDCAV technique for developing an autonomous driving navigation decision classification system using sensor information. The IBTAF-NDCAV method follows a sequence of steps in which the collected data are cleaned, checked for outliers, transformed, and normalized before further processing. After preprocessing, an improved MI-based FS method is used to select the most relevant features and reduce redundant information. A Bi-TACN is then used to learn temporal dependencies from the sensor data and classify the appropriate navigation decisions. The Adamax optimizer is also used to improve the training process and convergence. The performance of the IBTAF-NDCAV method is evaluated using the benchmark Autonomous Driving Sensor-Kinematics dataset, where it achieves an accuracy of 96.76% and performs better…

Editorial reading guide

  • What data created and tested the model, and was evaluation independent?
  • What is the cost of a false positive, false negative or out-of-distribution decision?