This study presents a cognitive driving model based on the Adaptive Control of Thought-Rational (ACT-R) architecture, integrated with the TORCS racing simulator via the Python library pytocl. The proposed model simulates human driving behavior and decision-making processes, focusing on critical highway-driving tasks such as lane-keeping, curve navigation, and adaptive braking to maintain safe following distances. A probabilistic visual attention mechanism, inspired by human drivers, dynamically allocates attention to different environmental regions, enhancing realistic behavioral simulations. The model showed sensitivity and adaptability to low traffic conditions. A controlled experiment was conducted with 21 human participants in order to evaluate the performance of the model in a highway scenario featuring symmetrical curves and low-density traffic flow. The experimental results demonstrate that the ACT-R model effectively replicates human driving behaviors with regard to speed and acceleration control. However, the model exhibits limitations in replicating the nuanced variability of human lateral control and attention shifts in complex scenarios. These findings underscore the potential of ACT-R in driving cognition modeling and identify areas for improvement, such as enhancing visual attention mechanisms and adaptive decision-making strategies. The study provides a foundation for advancing human-centered autonomous driving systems and adaptive driver-assistance technologies.
Underwater images frequently suffer from color casts, low illumination, and blur due to wavelength-dependent absorption and scattering. We present a practical two-stage, modular, and degradation-aware framework designed for real-time enhancement, prioritizing deployability on edge devices. Stage I employs a lightweight CNN to classify inputs into three dominant degradation classes (color cast, low light, blur) with 91.85% accuracy on an EUVP subset. Stage II applies three scene-specific lightweight enhancement pipelines and fuses their outputs using two alternative learnable modules: a global Linear Fusion and a LiteUNetFusion (spatially adaptive weighting with optional residual correction). Compared to the three single-scene optimizers (average PSNR = 19.0 dB; mean UCIQE ≈ 0.597; mean UIQM ≈ 2.07), the Linear Fusion improves PSNR by +2.6 dB on average and yields roughly +20.7% in UCIQE and +21.0% in UIQM, while maintaining low latency (~90 ms per 640 × 480 frame on an Intel i5-13400F (Intel Corporation, Santa Clara, CA, USA). The LiteUNetFusion further refines results: it raises PSNR by +1.5 dB over the Linear model (23.1 vs. 21.6 dB), brings modest perceptual gains (UCIQE from 0.72 to 0.74, UIQM 2.5 to 2.8) at a runtime of ≈125 ms per 640 × 480 frame, and better preserves local texture and color consistency in mixed-degradation scenes. We release implementation details for reproducibility and discuss limitations (e.g., occasional blur/noise amplification and domain generalization) together with future directions.
The frequent occurrence of traffic accidents highlights the importance of studying driving behavior to improve road safety. This study focused on the correlation between electroencephalogram (EEG) and driving braking, aiming to provide a basis for the improvement of autonomous driving technology. The experiment builds a highway scene based on the UC-Win/road driving simulation platform. The front vehicle has three driving modes: no braking, normal braking and emergency braking. The driver wears an Emotiv EPOC Flex Gel 32-channel EEG device to follow the front vehicle. The brake light of the previous car was turned on as the stimulation point, and the EEG signal of 1.2 seconds before and after the stimulation was extracted. Through event-related potential (ERP) analysis, the changes of N100, N200 and P300 latency amplitude topographic map and EEG frequency domain and time-frequency domain under different braking conditions were studied, and the correlation between EEG and braking behavior was discussed. The results showed that the N100, N200 and P300 components in the occipital lobe and frontal lobe changed significantly under different braking conditions. In the case of no braking, general braking and emergency braking, EEG signals are significantly different in Delta, Theta, Alpha and Beta bands; with the increase of braking degree, the brain activity shows the characteristics of high frequency oscillation near the braking stimulation point. The results of this study provide a reference for the driver’s intention recognition system, which is of great significance for understanding the correlation between EEG and driving behavior and the application of related technologies in the field of intelligent driving.
Driver workload classification requires a connection between observed multimodal responses and the cognitive demands of driving tasks. We propose ACT-R cognitive semantic lightweight fusion (ACTR-CSLF), which summarizes cognitive model states and production rules into 20 interpretable task semantic features and integrates them with 16 driving, eye-tracking, and electroencephalographic features. Evaluation covered 420 segments from 30 participants across seven simulated highway tasks and two weather conditions. In leave-one-subject-out validation, Human+ACT-R achieved Macro-F1 = 0.7277, compared with 0.5795 for Human-only and 0.7485 for ACT-R-only. Class-wise results showed improvements over Human-only at all three workload levels. Modality ablations identified input-dependent complementarity, while condition-level associations connected task semantics with subjective workload and physiological responses. Under simultaneous participant-and-scene holdout, pooled Macro-F1 was 0.5439 for Human+ACT-R, 0.4757 for ACT-R-only, and 0.4092 for Human-only; scene-averaged results characterized the uneven distribution of these gains. Condition baselines and random code controls separated task prior prediction from semantic interpretation. The joint classifier used 111 parameters, 1.0879 KB of storage, and 0.2147 ms mean single-thread CPU inference time. ACTR-CSLF provides a compact interface between explicit cognitive model knowledge and individual multimodal responses, combining workload discrimination with traceable task semantics.