• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.6
• /
• pp.313-330
• /
• 2021

In an autonomous driving system equipped with a camera and lidar, depth completion techniques enable dense depth estimation. In particular, using self-supervised learning it is possible to train the depth completion network even without ground truth. In actual autonomous driving, such depth completion should have very short latency as it is the input of other algorithms. So, rather than complicate the network structure to increase the accuracy like previous studies, this paper focuses on network latency. We design a U-Net type network with RegNet encoders optimized for GPU computation. Instead, this paper presents several techniques that can increase accuracy during the process of self-supervised learning. The proposed techniques increase the robustness to unreliable lidar inputs. Also, they improve the depth quality for edge and sky regions based on the semantic information extracted in advance. Our experiments confirm that our model is very lightweight (2.42 ms at 1280x480) but resistant to noise and has qualities close to the latest studies.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.55 no.6
• /
• pp.227-235
• /
• 2006

In this paper, autonomous separation methodology of faulted section based on network is proposed newly, which can minimize the outage effect as compared with the existing center-based faulted section separation method by determining and separating autonomously the faulted section by the free operation information exchange among IEDs on the feeder of distribution system. The all IEDs is designed in network in which client/server function is possible in order to separate autonomously the faulted section using PtP(Peer to Peer) communication. Also, Inference based solution of IED for the autonomous faulted section separation is designed by rules obtained from the analyzing results of distribution system topology. Here, the switch IEDs transmit on network the fault information utilizing on multi-casting communication method, at the fame time, determine selfly whether they operates or not by inferencing autonomously the faulted section using the inference-based solution after receiving the transmitted information. Finally, in order to verify the effectiveness and application possibility of the proposed methodology, the diversity fault cases are simulated for the typical distribution system.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.3
• /
• pp.374-383
• /
• 2004

This paper presents a neural network adaptive controller for autonomous diving control of an autonomous underwater vehicle (AUV) using adaptive backstepping method. In general, the dynamics of underwater robotics vehicles (URVs) are highly nonlinear and the hydrodynamic coefficients of vehicles are difficult to be accurately determined a priori because of variations of these coefficients with different operating conditions. In this paper, the smooth unknown dynamics of a vehicle is approximated by a neural network, and the remaining unstructured uncertainties, such as disturbances and unmodeled dynamics, are assumed to be unbounded, although they still satisfy certain growth conditions characterized by 'bounding functions' composed of known functions multiplied by unknown constants. Under certain relaxed assumptions pertaining to the control gain functions, the proposed control scheme can guarantee that all the signals in the closed-loop system satisfy to be uniformly ultimately bounded (UUB). Simulation studies are included to illustrate the effectiveness of the proposed control scheme, and some practical features of the control laws are also discussed.

• ETRI Journal
• /
• v.44 no.6
• /
• pp.1004-1019
• /
• 2022

For autonomous driving without high-definition maps, we present a model capable of generating multiple plausible paths from egocentric images for autonomous vehicles. Our generative model comprises two neural networks: feature extraction network (FEN) and path generation network (PGN). The FEN extracts meaningful features from an egocentric image, whereas the PGN generates multiple paths from the features, given a driving intention and speed. To ensure that the paths generated are plausible and consistent with the intention, we introduce an attentive discriminator and train it with the PGN under a generative adversarial network framework. Furthermore, we devise an interaction model between the positions in the paths and the intentions hidden in the positions and design a novel PGN architecture that reflects the interaction model for improving the accuracy and diversity of the generated paths. Finally, we introduce ETRIDriving, a dataset for autonomous driving, in which the recorded sensor data are labeled with discrete high-level driving actions, and demonstrate the state-of-the-art performance of the proposed model on ETRIDriving in terms of accuracy and diversity.

• Journal of information and communication convergence engineering
• /
• v.21 no.4
• /
• pp.346-350
• /
• 2023

Trajectory prediction is an essential element for driving autonomous vehicles, and various trajectory prediction models have emerged with the development of deep learning technology. Convolutional neural network (CNN) is the most commonly used neural network architecture for extracting the features of visual images, and the latest models exhibit high performances. This study was conducted to identify an efficient CNN backbone model among the components of deep learning models for trajectory prediction. We changed the existing CNN backbone network of multiple-trajectory prediction models used as feature extractors to various state-of-the-art CNN models. The experiment was conducted using nuScenes, which is a dataset used for the development of autonomous vehicles. The results of each model were compared using frequently used evaluation metrics for trajectory prediction. Analyzing the impact of the backbone can improve the performance of the trajectory prediction task. Investigating the influence of the backbone on multiple deep learning models can be a future challenge.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.10
• /
• pp.65-77
• /
• 2004

In this paper, we propose an adaptive mechanism based on immune algorithm and neural network identifier technique. It is also applied fur an autonomous guided vehicle (AGV) system. When the immune algorithm is applied to the PID controller, there exists the case that the plant is damaged due to the abrupt change of PID parameters since the parameters are almost adjusted randomly. To solve this problem, we use the neural network identifier (NNI) technique fur modeling the plant and humoral immune algorithm (HIA) which performs the parameter tuning of the considered model, respectively. After the PID parameters are determined in this off-line manner, these gains are then applied to the plant for the on-line control using an immune adaptive algorithm. Moreover, even though the neural network model may not be accurate enough initially, the weighting parameters are adjusted to be accurate through the on-line fine tuning. Finally, the simulation and experimental result fur the control of steering and speed of AGV system illustrate the validity of the proposed control scheme. These results for the proposed method also show that it has better performance than other conventional controller design methods.

• Molecules and Cells
• /
• v.26 no.4
• /
• pp.368-372
• /
• 2008

Recent molecular and genetic studies in rice, a short-day plant, have elucidated both conservation and divergence of photoperiod pathway genes and their regulators. However, the biological roles of rice genes that act within the autonomous pathway are still largely unknown. In order to better understand the function of the autonomous pathway genes in rice, we conducted molecular genetic analyses of OsFVE, a rice gene homologous to Arabidopsis FVE. OsFVE was found to be ubiquitously expressed in vegetative and reproductive organs. Overexpression of OsFVE could rescue the flowering time phenotype of the Arabidopsis fve mutants by up-regulating expression of the SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1) and down-regulating FLOWERING LOCUS C (FLC) expression. These results suggest that there may be a conserved function between OsFVE and FVE in the control of flowering time. However, OsFVE overexpression in the fve mutants did not rescue the flowering time phenotype in in relation to the response to intermittent cold treatment.

• Journal of Ocean Engineering and Technology
• /
• v.8 no.2
• /
• pp.131-140
• /
• 1994

Dynamics of AUV has heavy nonlinearities and many unknown parameters due to its bluff shape and low cruising speed. Intelligent algorithms, therefore, are required to overcome these nonlinearities and unknown system dynamics. Several identification techniques have been suggested for the application of control of underwater vehicles during last decade. This paper applies the neural network to identification and motion control problem of AUVs. Nonlinear dynamic systems of an AUV are identified using feedforward neural network. Simulation results show that the learned neural network can generate the motion of AUV. This paper, also, suggest an adaptive control scheme up-dates the controller weights with reference model and feedforward neural network using error back propagation.

• International Journal of Computer Science & Network Security
• /
• v.22 no.6
• /
• pp.139-144
• /
• 2022

Technological advancements taken the health care industry by a storm by embedding sensors in human body to measure their vitals. These smart solutions provide better and flexible health care to patients, and also easy monitoring for the medical practitioners. However, these innovative solutions provide their own set of challenges. The major challenge faced by embedding sensors in body is the issue of lack of infinite energy source. This work presents a meta-heuristic based routing model using modified PSO, and adopts an energy harvesting scheme to improve the network lifetime. The routing process is governed by modifying the fitness function of PSO to include charge, temperature and other vital factors required for node selection. A reactive routing model is adopted to ensure reliable packet delivery. Experiments have been performed and comparisons indicate that the proposed Energy Harvesting and Modified PSO (EHMP) model demonstrates low overhead, higher network lifetime and better network stability.

• Journal of Advanced Information Technology and Convergence
• /
• v.9 no.2
• /
• pp.55-63
• /
• 2019

Vehicle communication networks have some deficient network resources to support a vast multimedia service including safety driving information, video, news and some broadcast relayed from the playgrounds such as professional baseball games for autonomous vehicles. This paper deals with the framework for providing seamless multimedia service including safety driving information using multicast in cooperated-connected vehicle communication network (CVCN). It adopts smart-switch (SS) and smart intelligent multicast agent(SIMA) to support the seamless multimedia service. The SS manages and switches multimedia streams through SIMA in CVCN network. The SIMA to operate as an access point, is composed of multicast supporting part and control part of mobile devices/autonomous vehicles in CVCN network. Therefore this proposed technique using SS and SIMA within CVCN network is a new framework for multimedia service that can disperse the load of server.