This paper proposes the variable chromosome genetic algorithm (VCGA) for structure learning in neural networks. Currently, the structural parameters of neural networks, i.e., number of neurons, coupling relations, number of layers, etc., have mostly been designed on the basis of heuristic knowledge of an artificial intelligence (AI) expert. To overcome this limitation, in this study evolutionary approach (EA) has been utilized to automatically generate the proper artificial neural network (ANN) structures. VCGA has a new genetic operation called a chromosome attachment. By applying the VCGA, the initial ANN structures can be flexibly evolved toward the proper structure. The case study applied to the typical exclusive or (XOR) problem shows the feasibility of our methodology. Our approach is differentiated with others in that it uses a variable chromosome in the genetic algorithm. It makes a neural network structure vary naturally, both constructively and destructively. It has been shown that the XOR problem is successfully optimized using a VCGA with a chromosome attachment to learn the structure of neural networks. Research on the structure learning of more complex problems is the topic of our future research.

Research shows that urban intersections are a hotspot for traffic accidents which cause major human injuries. Predicting turning, passing, and stop maneuvers against surrounding vehicles is considered to be fundamental for advanced driver assistance systems (ADAS), or automated driving systems in urban intersections. In order to estimate the target intent in such situations, an interacting multiple model (IMM)-based intersection-target-intent estimation algorithm is proposed. A driver model is developed to represent the driver’s maneuvering on the intersection using an IMM-based target intent classification algorithm. The performance of the intersection-target-intent estimation algorithm is examined through simulation studies. It is demonstrated that the intention of a target vehicle is successfully predicted based on observations at an individual intersection by proposed algorithms.

This paper describes the development of deep learning based human-centered threat assessment for application to automated driving vehicle. To achieve naturalistic driver model that would feel natural while safe to a human driver, manual driving characteristics are investigated through real-world driving test data. A probabilistic threat assessment with predicted collision time and collision probability is conducted to evaluate driving situations. On the basis of collision risk analysis, two kinds of deep learning have been implemented to reflect human driving characteristics for automated driving. A deep neural network (DNN) and recurrent neural network (RNN) are designed by neural architecture search (NAS), and by learning from the sequential data, respectively. The NAS is used to automatically design the individual driver’s neural network for efficient and effortless design process while ensuring training performance. Sequential trends in the host vehicle’s state can be incorporated through hand-made RNN. It has been shown from human-centered risk assessment simulations that two successfully designed deep learning driver models can provide conservative and progressive driving behavior similar to a manual human driver in both acceleration and deceleration situations by preventing collision.

This paper proposes a human-centered risk assessment algorithm designed to find the intervention moment of drive mode and active safety mode while monitoring threat vehicles ahead to overcome effects of vehicular communication on risk assessment in automated driving vehicle. Although a conventional radar system is known to be best fitted on-board ranging sensor in terms of longitudinal safety, it is generally not enough for a reliable automated driving because of sensing uncertainty of the traffic environments and incomplete perception results due to sensor limitations. This can be overcome by implementing vehicle-to-vehicle (V2V) communication which provides complementary source of target vehicle’s dynamic behavior. Using V2V communication with vehicle internal and surround information obtained from the on-board sensor system, future vehicle motion has been predicted. With accurately predicted motion of a remote vehicle, a collision risk and the automated drive mode are determined by incorporating human factor. Effects of the V2V communication on a human-centered risk assessment algorithm have been investigated through a safe triangle analysis. The computer simulation studies have been conducted in order to validate the performance of the proposed algorithm. It has been shown that the V2V communication with the proposed risk assessment algorithm allows a faster drive mode decision and active safety intervention moment.

In this paper, we apply the neural architecture search (NAS) method to Korean grammaticality judgment tasks. Since the word order of a language is the final result of complex syntactic operations, a successful neural architecture search in linguistic data suggests that NAS can automate language model designing. Although NAS application to language has been suggested in the literature, we add a novel dataset that contains Korean-specific linguistic operations, which adds great complexity in the patterns. The result of the experiment suggests that NAS provides an architecture for the language. Interestingly, NAS has suggested an unprecedented structure that would not be designed manually. Research on the final topology of the architecture is the topic of our future research.

A genetic algorithm (GA) is a global search algorithm based on biological genetics. GAs are generally used for industrial applications, artificial neural networks, web applications, the defense industry, and so on. However, it is difficult to apply GAs to more complex situations because of the fixed number of chromosomes. In this research, in order to overcome this limitation, we propose a variable-chromosome GA with a chromosome attachment feature. Verification of the algorithm is carried out through anti-submarine high value unit (HVU) escort mission simulations. Ultimately, it is confirmed that the GA using the variable chromosome is more effective in dealing with highly complex missions, whereby the number of chromosomes gradually increases.