• 한국군사과학기술학회지
• /
• 제20권6호
• /
• pp.774-780
• /
• 2017

The techniques for driving trajectory calculation and driving trajectory distribution calculation are proposed to analyze the durability of ground vehicles effectively. To achieve this aim, the driving trajectory of a vehicle and the driving trajectory distribution of that are needed, in addition to road profile. The road profiles can be measured by a profilometer but a driving trajectory of a vehicle cannot be acquired effectively due to a large position error from a conventional GPS sensor. Therefore two techniques are proposed to reduce the position error of a vehicle and achieve the distribution of driving trajectory of that. The driving trajectory calculation technique produces relative positions by using the velocity, time and heading of a vehicle. The driving trajectory distribution calculation technique produces distributions of the driving trajectory by using axis transformation, estimating reference line, dividing sectors and plotting a histogram of the sectors. As a results of this study, we can achieve the considerably accurate driving trajectory and driving trajectory distribution of a vehicle.

• 동력기계공학회지
• /
• 제7권3호
• /
• pp.40-47
• /
• 2003

In this study, position and force simultaneous trajectory tracking control apparatus with pneumatic cylinder driving system is proposed. The pneumatic cylinder driving system that consists of two pneumatic cylinders constrained in series and two proportional flow control valves offers a considerable advantage as to non-interaction of the actuators because of the low stiffness of the pneumatic actuators. The controller applied to the driving system is composed of a non-interaction controller to compensate for interaction of two cylinders and a disturbance observer to reduce the effect of model discrepancy of the driving system in the low frequency range that cannot be suppressed by the non-interaction controller. The experimental results with the proposed control apparatus show that the interacting effects of two cylinders are eliminated remarkably and the proposed control apparatus tracks the given position and force trajectory accurately.

• 제어로봇시스템학회논문지
• /
• 제18권11호
• /
• pp.1065-1071
• /
• 2012

In this paper, we proposed a method for vehicle monitoring with trajectory reconstruction. For safety, it is important to monitor the driving habit of driver. Every year, many accidents occur due to the reckless driving of the driver. Continuous monitoring of the status of commercial vehicles is needed for safety through the entire path from start point to the destination. To monitor the reckless driving, we try to monitor the trajectory of the vehicle by using vehicle's lateral acceleration data. Compared with steering angle and lateral acceleration, these resemble each other. So, we find the relationship of steering angle and acceleration, and find the global direction of vehicle. We find the position of non-GPS section with EKF (External Kalman Filter) and reconstruct the whole trajectory during vehicle driving.

• 동력기계공학회지
• /
• 제10권3호
• /
• pp.97-103
• /
• 2006

In this study, a position trajectory tracking control algorithm is proposed for a pneumatic artificial muscle driving apparatus composed of a actuator which imitates the muscle of human, a position sensor and a control valve. The controller applied to the driving apparatus is composed of a state feedback controller and disturbance observer. The feedback controller which feeds back position, velocity and acceleration is derived from the linear model of pneumatic artificial muscle driving apparatus. The disturbance observer is designed to improve trajectory tracking performance and to reduce the effect of model discrepancy. The effectiveness of the designed controller is proved by experiments and the experimental results show that the pneumatic artificial muscle driving apparatus with the proposed control algorithm tracks given position reference inputs accurately.

• 제어로봇시스템학회논문지
• /
• 제19권7호
• /
• pp.599-604
• /
• 2013

Path planning of mobile robots has a purpose to design an optimal path from an initial position to a target point. Minimum driving time, minimum driving distance and minimum driving error might be considered in choosing the optimal path and are correlated to each other. In this paper, an efficient driving trajectory is planned in a real situation where a mobile robot follows a moving object. Position and distance of the moving object are obtained using a web camera, and the rotation angular and linear velocities are estimated using Kalman filters to predict the trajectory of the moving object. Finally, the mobile robot follows the moving object using a single curvature trajectory by estimating the trajectory of the moving object. Using the estimation by Kalman filters and the single curvature in the trajectory planning, the total tracking distance and time saved amounts to about 7%. The effectiveness of the proposed algorithm has been verified through real tracking experiments.

• Journal of Mechanical Science and Technology
• /
• 제19권5호
• /
• pp.1107-1115
• /
• 2005

In this study, a position and force simultaneous trajectory tracking control algorithm is proposed for a driving apparatus that consists of two pneumatic cylinders connected in series. The controller applied to the driving apparatus is composed of a non-interaction controller to compensate for interaction between cylinders and a disturbance observer aimed to reduce the effect of model discrepancy that cannot be compensated by the non-interaction controller. The effectiveness of the proposed control algorithm is proved by experimental results.

• ETRI Journal
• /
• 제37권5호
• /
• pp.1032-1043
• /
• 2015

ETRI's Co-Pilot project is aimed at the development of an automated vehicle that cooperates with a driver and interacts with other vehicles on the road while obeying traffic rules without collisions. This paper presents a core block within the Co-Pilot system; the block is named "Co-Pilot agent" and consists of several main modules, such as road map generation, decision-making, and trajectory generation. The road map generation builds road map data to provide enhanced and detailed map data. The decision-making, designed to serve situation assessment and behavior planning, evaluates a collision risk of traffic situations and determines maneuvers to follow a global path as well as to avoid collisions. The trajectory generation generates a trajectory to achieve the given maneuver by the decision-making module. The system is implemented in an open-source robot operating system to provide a reusable, hardware-independent software platform; it is then tested on a closed road with other vehicles in several scenarios similar to real road environments to verify that it works properly for cooperative driving with a driver and automated driving.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권6호
• /
• pp.1562-1582
• /
• 2024

The prediction of pedestrian trajectory is conducive to reducing traffic accidents and protecting pedestrian safety, which is crucial to the task of intelligent driving. The existing methods mainly use the past pedestrian trajectory to predict the future deterministic pedestrian trajectory, ignoring pedestrian intention and trajectory diversity. This paper proposes a multi-modal trajectory prediction model that introduces pedestrian intention. Unlike previous work, our model makes multi-modal goal-conditioned trajectory pedestrian prediction based on the past pedestrian trajectory and pedestrian intention. At the same time, we propose a novel Gate Recurrent Unit (GRU) to process intention information dynamically. Compared with traditional GRU, our GRU adds an intention unit and an intention gate, in which the intention unit is used to dynamically process pedestrian intention, and the intention gate is used to control the intensity of intention information. The experimental results on two first-person traffic datasets (JAAD and PIE) show that our model is superior to the most advanced methods (Improved by 30.4% on MSE_0.5s and 9.8% on MSE_1.5s for the PIE dataset; Improved by 15.8% on MSE_0.5s and 13.5% on MSE_1.5s for the JAAD dataset). Our multi-modal trajectory prediction model combines pedestrian intention that varies at each prediction time step and can more comprehensively consider the diversity of pedestrian trajectories. Our method, validated through experiments, proves to be highly effective in pedestrian trajectory prediction tasks, contributing to improving traffic safety and the reliability of intelligent driving systems.

• 대한임베디드공학회논문지
• /
• 제19권2호
• /
• pp.101-106
• /
• 2024

For the commercialization of automated vehicles, it is necessary to create various scenarios that can evaluate driving safety and establish a data system that can verify them. Depending on the vehicle's ODD (Operational Design Domain), there are numerous scenarios with various parameters indicating vehicle driving conditions, but no systematic methodology has been proposed to create and combine scenarios to test them. Therefore, projects are actively underway abroad to establish a scenario library for real-world testing or simulation of autonomous vehicles. However, since it is difficult to obtain data, research is being conducted based on simulations that simulate real road. Therefore, in this study, parameters calculated through individual vehicle trajectory data extracted based on roadside CCTV image-based driving environment DB was proposed through the extracted data. This study can be used as basic data for safety standards for scenarios representing various driving behaviors.

• Journal of information and communication convergence engineering
• /
• 제21권4호
• /
• 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.