• ETRI Journal
• /
• v.45 no.5
• /
• pp.795-810
• /
• 2023

Unmanned aerial vehicles (UAV) and ground vehicles (UGV) require advanced video analytics for various tasks, such as moving object detection and segmentation; this has led to increasing demands for these methods. We propose a zero-shot video object segmentation method specifically designed for UAV and UGV applications that focuses on the discovery of moving objects in challenging scenarios. This method employs a background memory model that enables training from sparse annotations along the time axis, utilizing temporal modeling of the background to detect moving objects effectively. The proposed method addresses the limitations of the existing state-of-the-art methods for detecting salient objects within images, regardless of their movements. In particular, our method achieved mean J and F values of 82.7 and 81.2 on the DAVIS'16, respectively. We also conducted extensive ablation studies that highlighted the contributions of various input compositions and combinations of datasets used for training. In future developments, we will integrate the proposed method with additional systems, such as tracking and obstacle avoidance functionalities.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.25 no.1
• /
• pp.99-107
• /
• 2014

Abruptly occurred obstacles on highway threaten driving safety. Radar draws the attention to the collision avoidance system because it can be fully operational in all weather, and day and night condition. This paper presents the design, implementation and performance test results of pulsed Doppler radar system for detection and warning of road obstacles. The system is designed to consider highway environment and detection capability about various fixed and moving obstacles. The system consists of 4 subsystems, which include antenna unit, transmitter and receiver unit, radar signal & data processing unit, and controller & display unit. The core technologies include clutter map based change detection for fixed obstacles detection, Doppler estimation for velocity detection of moving targets, and azimuth angle estimation method using monopulse for lane estimation and tracking. The design performance of the developed radar system is verified through experiments using a fixed reference target and moving vehicles in test highway.

• Journal of Korea Game Society
• /
• v.6 no.3
• /
• pp.3-12
• /
• 2006

Path-finding, one of the traditional Game A.I. problems, becomes an important issue to make games more realistic. Due to the limited resources in the computer system, path-finding systems sometimes produce a simplified and unrealistic path. The most relent researches have been focused on the path-finding avoiding only static obstacles. Various moving obstacles are however deployed in real games, a method avoiding those obstacles and producing a smooth path is necessary. In this paper, navigation mesh is used to represent 3D space and its topological characteristics are used for path-finding. Intellectual repulser and attractor are also used to avoid moving obstacles and to find an optimal path. We have evaluated the path produced by the method proposed in this paper and verified its usability in real game.

• 전자공학회논문지 IE
• /
• v.46 no.3
• /
• pp.33-41
• /
• 2009

Because the stability of obstacle avoidance ability is important for the safe operation of mobile robots, this paper deals with the analysis of dynamic stability of Limit-cycle navigation method that was proposed by authors. Limit-cycle navigation method is fast and easy to implement for fast moving mobile robots using limit-cycle characteristics of the 2nd-order nonlinear function. It can be applied to robots in dynamically changing environment such as the robot soccer. By adjusting the radius of the motion circle and the direction of the obstacle avoidance, the mobile robot can avoid the collision with obstacles and move to the destination point. The stability of Limit-cycle navigation method is analyzed with a linear model. To demonstrate the effectiveness and applicability, it is applied to the robot soccer Simulations and real experiments ascertain the merits of the proposed method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.10
• /
• pp.939-947
• /
• 2010

With the rising numbers of elderly and disabled people, the demand for welfare services using a robotic system and not involving human effort is likewise increasing. This study deals with a mobile-robot system combined with a BWS (Body Weight Support) system for gait rehabilitation. The BWS system is designed via the kinematic analysis of the robot's body-lifting characteristics and of the walking guide system that controls the total rehabilitation system integrated in the mobile robot. This mobile platform is operated by utilizing the AGV (Autonomous Guided Vehicle) driving algorithm. Especially, the method that integrates geometric path tracking and obstacle avoidance for a nonholonomic mobile robot is applied so that the system can be operated in an area where the elderly users are expected to be situated, such as in a public hospital or a rehabilitation center. The mobile robot follows the path by moving through the turning radius supplied by the pure-pursuit method which is one of the existing geometric path-tracking methods. The effectiveness of the proposed method is verified through the real experiments those are conducted for path tracking with static- and dynamic-obstacle avoidance. Finally, through the EMG (Electromyography) signal measurement of the subject, the performance of the proposed system in a real operation condition is evaluated.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.2
• /
• pp.161-166
• /
• 2006

Localization of mobile agent within a sensing network is a fundamental requirement for many applications, using networked navigating systems such as the sonar-sensing system or the visual-sensing system. To fully utilize the strengths of both the sonar and visual sensing systems, This paper describes a networked sensor-based navigation method in an indoor environment for an autonomous mobile robot which can navigate and avoid obstacle. In this method, the self-localization of the robot is done with a model-based vision system using networked sensors, and nonstop navigation is realized by a Kalman filter-based STSF(Space and Time Sensor Fusion) method. Stationary obstacles and moving obstacles are avoided with networked sensor data such as CCD camera and sonar ring. We will report on experiments in a hallway using the Pioneer-DX robot. In addition to that, the localization has inevitable uncertainties in the features and in the robot position estimation. Kalman filter scheme is used for the estimation of the mobile robot localization. And Extensive experiments with a robot and a sensor network confirm the validity of the approach.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.5
• /
• pp.405-411
• /
• 2011

This paper presents a hierarchical trajectory planning method which can handle a collision-free of the planned path in complex and dynamic environments. A PV (Path & Velocity profile) planning method minimizes a sharp change of orientation and waiting time to avoid a collision with moving obstacle through detour path. The path generation problem is solved by three steps. In the first step, a smooth global path is generated using $A^*$ algorithm. The second step sets up the velocity profile for the optimization problem considering the maximum velocity and acceleration. In the third step, the velocity profile for obtaining the shortest path is optimized using the fuzzy and genetic algorithm. To show the validity and effectiveness of the proposed method, realistic simulations are performed.

• Journal of IKEEE
• /
• v.25 no.2
• /
• pp.337-343
• /
• 2021

Deep reinforcement learning is an artificial intelligence algorithm that enables learners to select optimal behavior based on raw and, high-dimensional input data. A lot of research using this is being conducted to create an optimal movement path of a mobile robot in an environment in which obstacles exist. In this paper, we selected the Dueling Double DQN (D3QN) algorithm that uses the prioritized experience replay to create the moving path of mobile robot from the image of the complex surrounding environment. The virtual environment is implemented using Webots, a robot simulator, and through simulation, it is confirmed that the mobile robot grasped the position of the obstacle in real time and avoided it to reach the destination.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.6
• /
• pp.549-559
• /
• 2015

In this paper, an efficient dynamic reactive motion planning method for an autonomous vehicle in a dynamic environment is proposed. The purpose of the proposed method is to improve the robustness of autonomous robot motion planning capabilities within dynamic, uncertain environments by integrating a virtual plane-based reactive motion planning technique with a sensor fusion-based obstacle detection approach. The dynamic reactive motion planning method assumes a local observer in the virtual plane, which allows the effective transformation of complex dynamic planning problems into simple stationary ones proving the speed and orientation information between the robot and obstacles. In addition, the sensor fusion-based obstacle detection technique allows the pose estimation of moving obstacles using a Kinect sensor and sonar sensors, thus improving the accuracy and robustness of the reactive motion planning approach. The performance of the proposed method was demonstrated through not only simulation studies but also field experiments using multiple moving obstacles in hostile dynamic environments.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.2
• /
• pp.918-925
• /
• 2017

This paper presents a path planning problem for multi-robot system in the environment with dynamic obstacles. In order to guide the robots move along a collision-free path efficiently and reach the goal position quickly, a navigation method based on fuzzy logic controllers has been developed by using proximity sensors. There are two kinds of fuzzy controllers developed in this work, one is used for obstacle avoidance and the other is used for orientation to the target. Both static and dynamic obstacles are included in the environment and the dynamic obstacles are defined with no type of restriction of direction and velocity. Here, the environment is unknown for all the robots and the robots should detect the surrounding information only by the sensors installed on their bodies. The simulation results show that the proposed method has a positive effectiveness for the path planning problem.