No abstract

This paper presents a novel development of an ecological (eco) driving system for running a vehicle on roads with up-down slopes. Fuel consumed in a vehicle is greatly influenced by road gradients, aside from its velocity and acceleration characteristics. Therefore, optimum control inputs can only be computed through anticipated rigorous reasoning using information concerning road terrain, model of the vehicle dynamics, and fuel consumption characteristics. In this development, a nonlinear model predictive control method with a fast optimization algorithm is implemented to derive the vehicle control inputs based on road gradient conditions obtained from digital road maps. The fuel consumption model of a typical vehicle is formulated using engine efficiency characteristics and used in the objective function to ensure fuel economy driving. The proposed eco-driving system is simulated on a typical road with various shapes of up-down slopes. Simulation results reveal the ability of the eco-driving system in significantly reducing fuel consumption of a vehicle. The fuel saving behavior is graphically illustrated, compared, and analyzed to focus on the significance of this development.

This paper presents a coordination scheme of automated vehicles at an intersection without using any traffic lights. Using a two-way communication network, vehicles approaching the intersection from all sections are globally coordinated, by considering their states all together in a model predictive control framework, in order to achieve smooth traffic flows at the intersection. The optimal trajectories of the vehicles are computed based on avoidance of their cross-collision risks around the intersection under relevant constraints and preferences. The scheme efficiently utilizes the intersection area by preventing each pair of conflicting vehicles from approaching their cross-collision point at the same time, instead of reserving the whole intersection area for the conflicting vehicles one after another. The scheme also enables leftor right-turning movements of vehicles under constrained velocity without using any auxiliary lanes. The proposed vehicle-intersection coordination scheme is evaluated through numerical simulation in a typical test intersection consisting of both multilanes and single-lane approaches with turning movements of vehicles. Observations under different traffic flow conditions reveal that the proposed scheme significantly improves intersection performance compared with the traditional signalized intersection scheme.

Human facial emotion recognition (FER) has attracted the attention of the research community for its promising applications. Mapping different facial expressions to the respective emotional states are the main task in FER. The classical FER consists of two major steps: feature extraction and emotion recognition. Currently, the Deep Neural Networks, especially the Convolutional Neural Network (CNN), is widely used in FER by virtue of its inherent feature extraction mechanism from images. Several works have been reported on CNN with only a few layers to resolve FER problems. However, standard shallow CNNs with straightforward learning schemes have limited feature extraction capability to capture emotion information from high-resolution images. A notable drawback of the most existing methods is that they consider only the frontal images (i.e., ignore profile views for convenience), although the profile views taken from different angles are important for a practical FER system. For developing a highly accurate FER system, this study proposes a very Deep CNN (DCNN) modeling through Transfer Learning (TL) technique where a pre-trained DCNN model is adopted by replacing its dense upper layer(s) compatible with FER, and the model is fine-tuned with facial emotion data. A novel pipeline strategy is introduced, where the training of the dense layer(s) is followed by tuning each of the pre-trained DCNN blocks successively that has led to gradual improvement of the accuracy of FER to a higher level. The proposed FER system is verified on eight different pre-trained DCNN models (VGG-16, VGG-19, ResNet-18, ResNet-34, ResNet-50, ResNet-152, Inception-v3 and DenseNet-161) and well-known KDEF and JAFFE facial image datasets. FER is very challenging even for frontal views alone. FER on the KDEF dataset poses further challenges due to the diversity of images with different profile views together with frontal views. The proposed method achieved remarkable accuracy on both datasets with pre-trained models. On a 10-fold cross-validation way, the best achieved FER accuracies with DenseNet-161 on test sets of KDEF and JAFFE are 96.51% and 99.52%, respectively. The evaluation results reveal the superiority of the proposed FER system over the existing ones regarding emotion detection accuracy. Moreover, the achieved performance on the KDEF dataset with profile views is promising as it clearly demonstrates the required proficiency for real-life applications.