• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.42 no.10
• /
• pp.842-850
• /
• 2014

This paper presents a trajectory tracking controller for rotorcraft UAVs to improve the tracking performances in the presence of various uncertainties. The proposed tracking method consists of a velocity guidance law based on the relative distance and L1 adaptive augmentation loop for tracking the velocity commands. In the proposed structure, the desired velocity generated by the guidance law is the reference value of the adaptive controller for accurate path tracking. In the guidance law, the desired acceleration is generated based on the relative distance and its derivatives, and then the velocity command of the inner control loop is calculated by integrating the accelerations. $L_1$ augmentation loop supplements the linear controller to guarantee the flight performances such as a tracking accuracy in the presence of the uncertainties. The proposed controller was validated in actual flight tests to successfully demonstrate its capability using a quadrotor UAV.

• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.6
• /
• pp.53-61
• /
• 2014

For monitoring the flight trajectory and the status of a launch vehicle, the mission control system in NARO space center process data acquired from the ground tracking system, which consists of two tracking radars, four telemetry stations, and one electro-optical tracking system. Each tracking unit exhibits its own tracking error mainly due to multi-path, clutter and radio refraction, and by utilizing only one among transmitted informations, it is not possible to determine the actual vehicle trajectory. This paper presents a way of generating flight trajectory via post-processing the data received from the ground tracking system. The post-processing algorithm is divided into two parts: compensation for atmosphere radio refraction and multi-sensor fusion, for which a decentralized Kalman filter was adopted and implemented based on constant acceleration model. Applications of the present scheme to real data resulted in the flight trajectory where the tracking errors were minimized than done by any one sensor.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.3-12
• /
• 2023

In the post-COVID era, the importance of quarantine measures is greatly emphasized, and accordingly, research related to the detection of mask wearing conditions and prevention of other infectious diseases using deep learning is being conducted. However, research on the detection and tracking of visitors to cultural facilities to prevent the spread of diseases is equally important, so research on this should be conducted. In this paper, a convolutional neural network-based object detection model is trained through transfer learning using a pre-collected dataset. The weights of the trained detection model are then applied to a multi-object tracking model to monitor visitors. The visitor detection model demonstrates results with a precision of 96.3%, recall of 85.2%, and an F1-score of 90.4%. Quantitative results of the tracking model include a MOTA (Multiple Object Tracking Accuracy) of 65.6%, IDF1 (ID F1 Score) of 68.3%, and HOTA (Higher Order Tracking Accuracy) of 57.2%. Furthermore, a qualitative comparison with other multi-object tracking models showcased superior results for the model proposed in this paper. The research of this paper can be applied to the hygiene systems within cultural facilities in the post-COVID era.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.12
• /
• pp.583-595
• /
• 2002

Homing operation can be defined as a series of actions which are necessary for a mobile robot to move from the current position with any arbitrary orientation to a desired position with a specified orientation, while avoiding possible obstacles. In this paper, a homing and obstacle avoidance algorithm for nonholonomic mobile robots is proposed. The proposed algorithm consists of a local goal generator, a discrete state controller, and local path tracking controller based on Aicardi's path following algorithm. In the discrete state controller, 4 states are defined according to the environmental conditions and 4 desired high-level command for the states are given as follows: avoid, wander, home and homing zones. The proposed local goal generator is designed to generate the desired local path by using weighted distance transforms which are newly made to satisfy the nonholonomic constraints of mobile robots. Here, subgoals are also found as vertices of the desired local path. To demonstrate result effectiveness and applicability of the proposed algorithm, computer simulations are illustrated and experimental results for a real mobile robot system are also provided.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.1
• /
• pp.45-55
• /
• 1998

In this paper, a robust path tracker and depth controller of Autonomous Underwater Vehicle based on sliding mode control is presented. We have also designed augmented equivalent control inputs by analyzing the sliding mode with the reaching mode. This can enhance the reaching rate, and improve chattering problems, that is, noise caused by the control plane actuator of the vehicle, which is one of the problems that occur when sliding mode control is used. Also to resolve the steady state error generated in the path tracker under current effect, a modified sliding plane is constructed. Also a redesigned sliding plane and control input using transformation matrix is proposed to do easy design of MIMO depth controller. For state variables that cannot be measured directly, reduced order sliding mode control is used to design an observer. The performance of designed path tracker and depth controller is investigated by computer simulation. The results show that the proposed control system has robust performance to parameter variation, modelling error and disturbance.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.472-477
• /
• 2023

In this paper, we proposes the use of Model Predictive Control (MPC) techniques to enable quadcopter drones to effectively follow paths and maintain flight safety even under dynamic external environments and disturbances. Through simulations conducted in MATLAB/Simulink, the performance of two controllers, PID and MPC, is compared in flight scenarios with disturbances. The proposed design method shows that the MPC controller, when compared to the PID controller, exhibits a difference in the Mean Squared Error between the intended flight path and the actual path of the quadcopter drone. This difference is 0.2 in performance under no disturbance, and it increases to 0.8 under disturbance, demonstrating the improved path following accuracy of the MPC controller.

• Journal of Institute of Control, Robotics and Systems
• /
• v.19 no.8
• /
• pp.745-754
• /
• 2013

An efficient urban outdoor localization and navigation system is proposed for car-like robots. First an accurate outdoor localization method is suggested using line/arc features and 2.5D map matching with LRFs (Laser Range Finders), which can reduce the number of singular cases and increase accuracy. Also, path generation, path tracking, and path modification algorithms are proposed for navigation. All these algorithms are implemented on an electric scooter to construct an autonomous urban outdoor localization and navigation system. Experiments reveal the practicality of the proposed system.

• Journal of Drive and Control
• /
• v.17 no.1
• /
• pp.13-20
• /
• 2020

Adaptive neural networks based lateral controller is presented to guarantee path following performance for vehicle lane keeping in the presence of parameter time-varying characteristics of the vehicle lateral dynamics due to the road surface condition, load distribution, tire pressure and so on. The proposed adaptive controller could compensate vehicle lateral dynamics deviated from nominal dynamics resulting from parameter variations by incorporating it with neural networks that have the ability to approximate any given nonlinear function by adjusting weighting matrices. The controller is derived by using Lyapunov-based approach, which provides adaptive update rules for weighting matrices of neural networks. To show the superiority of the presented adaptive neural networks controller, the simulation results are given while comparing with backstepping controller chosen as the baseline controller. According to the simulation results, it is shown that the proposed controller can effectively keep the vehicle tracking the pre-given trajectory in high velocity and curvature with much accuracy under parameter variations.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.8
• /
• pp.948-954
• /
• 1999

In this paper, we propose a cooperative motion planning algorithm for two tightly-coupled mobile robots. Specifically, the considered cooperative work is that two mobile robots should transfer a long rigid object along a predefined path. To resolve the problem, we introduce a master-slave concept for two obile robots having the same structure. According to the velocity of the master robot and the positions of two robots on the path, the velocity of the slave robot is determined. The slave normally tracks the master's motion, but in case that the velocity of the slave exceeds the velocity limit, the roles of the robots should be interchanged. The effectiveness of the proposed algorithm is proved by computer simulations.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.120.4-120
• /
• 2001

In this paper, a method of generating trajectories in real time for a mobile robot in a dynamic environment is proposed. Specifically, this method is focused on soccer-playing robots that need to calculate trajectories in real time, which are constantly subject to rapidly change as targets and obstacles move. The robots also should move at the fastest available speed, while tracking the generated trajectories. The method proposed in this paper solves the geometric problem of finding a smooth curve that joins two endpoints. To have this solved, we assign five constraints to each endpoint, which are the usual x, y, theta, and curvature as well as the influence of the initial robot velocity on the path. With these five constraints, the path generated can always be physically followed by robot. Through this method, the travel time of the robot over the entire path can b optimized. Therefore it can ...