• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2000년도 추계학술대회논문집A
• /
• pp.647-652
• /
• 2000

This paper proposes a collision warning/avoidance algorithm using a trajectory prediction method. This algorithm is based on 2-dimensional kinematics and the Kalman filter has been used to obtain the information of the object vehicle. This algorithm has been investigated via computer simulation and showed a good trajectory prediction performance. The proposed collision warning/avoidance algorithm would enhanced driver acceptance for a collision warning/avoidance system.

• 한국자동차공학회논문집
• /
• 제7권8호
• /
• pp.233-247
• /
• 1999

Traffic accidents are normally caused by late or faulty judgements due to the driver's inaccurate estimation of the distance, velocity, and acceleration from the surrounding vehicles as well as his carelessness or inattention. Thus, the development of collision avoidance systems is motivated by their great potential for increased vehicle safety. A typical collision avoidance system consists of the forward-looking sensor, the criteria for activation of collision warming and avoidance, the collision avoidance maneuvers, and the user interface. This thesis is concerned with the development of a collision warning algorithm in which the driver is warned of approaching collision with the visual and/or the audible signals . The warning algorithm based on fuzzy logic is presented here based on new warning criteria. It has been newly derived from the conventional warning equation by adding a new input variable of the required deceleration to avoid collision. The algorithm is also able to adapt to the individual driver's taste along with the different road conditions by externally controlling the warning intensity. Finally , the proposed algorithm has been validated using computer simulation.

• 자동차안전학회지
• /
• 제5권1호
• /
• pp.5-10
• /
• 2013

This paper describes a Forward Collision Warning (FCW) system based on the radar driven fusion with camera. The objective of FCW system is to provide an appropriate alert with satisfying the evaluation scenarios of US-NCAP and a driver acceptance. For this purpose, this paper proposed a data fusion algorithm and a collision warning algorithm. The data fusion algorithm generates information of fusion target depending on the confidence of camera sensor. The collision warning algorithm calculates indexes and determines an appropriate alert-timing by using analysis results of manual driving data. The FCW system with the proposed data fusion and collision warning algorithm was investigated via scenarios of US-NCAP and a real-road driving. It is shown that the proposed FCW system can improve the accuracy of an alarm-timing and reduce the false alarm in real roads.

• 한국자동차공학회논문집
• /
• 제6권5호
• /
• pp.222-227
• /
• 1998

Collision warning systems have been an active research and development area as the interests and demands for ASV's (Advanced Safety Vehicles) have increased. This paper presents an experimental investigation of a collision warning system for automobiles. A collision warning HiLS(Hardware-in-the-Loop Simulation) system has been designed and used to test the collision warning algorithm, radar sensors, and warning displays under realistic operating conditions in the laboratory. the collision warning algorithm is operated by a warning index, which is a function of the warning distance and the braking distance. The computer calculates velocities of the preceding vehicle and following vehicle, relative distance and relative velocity of the vehicles using vehicle simulation models. The relative distance and the relative velocity are applied to the vehicle simulator controlled by a DC motor.

• 대한임베디드공학회논문지
• /
• 제6권3호
• /
• pp.157-167
• /
• 2011

Although vehicles become more intelligent for convenience and safety of drivers, traffic accidents are increased more and more. Especially, car-to-car single rear impacts in the urban area are increased rapidly because of driver inattention. To prevent rear impacts in the urban area, commercial automobile vendor applies the low-speed short-distance collision warning system. This paper presents low-speed short-distance collision warning algorithm for the city driving by using sensor fusion of laser sensor and ultrasonic sensor. An experiment using embedded microprocessor in the driving track was used to demonstrate the feasibility of the collision warning algorithm.

• Tribology and Lubricants
• /
• 제30권3호
• /
• pp.168-176
• /
• 2014

This paper presents an engine oil change warning algorithm based on the test results of a small dip-stick-gage-type engine-oil-deterioration-detection sensor, software to realize the algorithm and a display device to apply the software. The algorithm determines the engine oil deterioration condition from the rate of change in the dielectric constant based on the average measured capacitance at $80^{\circ}C$ after the engine stops. The rate of change in the dielectric constant at the time for oil change correlates with the time that one of recommended warning limits for engine oil physical properties such as TAN (Total Acid Number), TBN (Total Base Number) and viscosity is first reached. At this point, a warning signal for oil change appears on the display device like a smart-phone or individual display device. The frames of smartphone app have three stages. The user can directly input all of the thresholds into the frame of the smartphone app. The screen of the display device comprises one frame for each warning signature with the related message. The user can input the thresholds to the device through a USB cable connected to a personal computer.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.656-659
• /
• 1997

In order to improve vehicle safety, collision warning systems have been proposed by many researchers. This paper presents several algorithms to determine the degree of real end collision by using fuzzy logic and neural networks. In order to provide realistic data for the algorithm design, a data collection system has been installed on a passenger car.

• 한국군사과학기술학회지
• /
• 제14권3호
• /
• pp.474-481
• /
• 2011

The Electronic Warfare Computer for the Aircraft Survivability Equipment will improve the ability for countermeasures by analysis about threat information. This paper suggests method that threat data integration of multiple sensors(Radar Warning Receiver, Laser Warning Receiver, Missile Warning Receiver). The algorithm of threat data integration is based on detected threat sequence and azimuth information. The threat sequence information is analyzed in advance and the azimuth data is received from sensors. The suggested method is evaluated through simulation under the environment like real helicopter.

• Journal of Information Processing Systems
• /
• 제20권2호
• /
• pp.263-272
• /
• 2024

A model based on genetic algorithm optimization, GA-SVM, is proposed to warn university students of their status. This model improves the predictive effect of support vector machines. The genetic optimization algorithm is used to train the hyperparameters and adjust the kernel parameters, kernel penalty factor C, and gamma to optimize the support vector machine model, which can rapidly achieve convergence to obtain the optimal solution. The experimental model was trained on open-source datasets and validated through comparisons with random forest, backpropagation neural network, and GA-SVM models. The test results show that the genetic algorithm-optimized radial basis kernel support vector machine model GA-SVM can obtain higher accuracy rates when used for early warning in university learning.

• 한국군사과학기술학회지
• /
• 제11권1호
• /
• pp.93-101
• /
• 2008

It is difficult for a MWR(Missile Warning Radar) to perform a threat decision accurately since there is no tracking part which gives more accurate threat information to the MWR. In this paper, the threat decision algorithm is proposed using an azimuth angular rate to improve the accuracy. The azimuth angular rate is dependent upon the direction of an approaching target. The target is classified into a threat or non-threat using a boundary condition of the azimuth angular rate. The boundary condition is determined using the Monte-Carlo simulation. The performance of the proposed algorithm is evaluated using this condition at field tests of MWR. The efficiency of the proposed method for the threat decision is proved by comparing the results of field tests with the simulation results.