• 로봇학회논문지
• /
• 제12권3호
• /
• pp.258-262
• /
• 2017

In this paper, we propose a new collision detection algorithm for human-robot collaboration. We use an IMU sensor located at the tip of the manipulator and the kinematic behavior of the manipulator to detect the unexpected collision between the robotic manipulator and environment. Unlike other method, the developed algorithm uses only the kinematic relationship between the manipulator joint and the end effector. Therefore, the collision estimation signal is not affected by the error of the dynamics model. The proposed collision detection algorithm detects the collision by comparing the estimated acceleration of the end effector derived from the position, velocity and acceleration trajectories of the robot joints with the actual acceleration measured by the sensor. In simulation, we compare the performance of our method with the conventional Residual Observer (ROB). Our method is less sensitive to the load variation because of the independency on the dynamic modeling of the manipulator.

• 제어로봇시스템학회논문지
• /
• 제21권12호
• /
• pp.1089-1099
• /
• 2015

This paper proposes vision-based collision risk estimation method for an unmanned surface vehicle. A robust image-processing algorithm is suggested to detect target obstacles from the vision sensor. Vision-based Target Motion Analysis (TMA) was performed to transform visual information to target motion information. In vision-based TMA, a camera model and optical flow are adopted. Collision risk was calculated by using a fuzzy estimator that uses target motion information and vision information as input variables. To validate the suggested collision risk estimation method, an unmanned surface vehicle experiment was performed.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2013년도 추계학술대회
• /
• pp.387-389
• /
• 2013

In order to determine the collision risk, the probability estimation is very important part for accurate risk estimation. Recently, the collision risk at the Busan North Port is studied for making the risk map by authors. The result has been found some connections with previous collision places. For more precise estimation, the probability calculation is necessary. Recently the Bayesian matrix is mainly used for calculating the probabilities. Also considering the oil spill risk with tankers, ships' speed, relative angle and ships' size are key aspect whether breaking the double hull or not. This research presents the way of estimating the probabilities not her research and also the collision risk probability considerations for small divided areas.

• 한국통신학회논문지
• /
• 제36권3B호
• /
• pp.248-253
• /
• 2011

선박의 충돌회피 알고리즘은 인적 요인에 의한 해난사고를 방지하고, 보다 효과적이고 안전한 운항을 위해 개발되어 왔다. 본 논문에서는 충돌사고의 위험성을 줄이고, 안전운항을 지원하기 위하여, 베이지안 추정 이론을 이용하여 충돌 위험도를 추정하는 알고리즘을 고안한다. 기존의 선박 충돌 회피를 시스템보다 안전한 충돌 회피를 위해서 베이지안 추정 이론을 이용하여 충돌 위험도를 계산하고 일정 시간 이후에 타선들의 위치 및 속도정보와 자선의 위치 및 속도정보를 이용하여 충돌 위험도를 예측함으로써 보다 안전하고 효율적인 충돌 위험도를 결정한다. 본 논문에서 타선박의 항해정보는 AIS 정보를 가정하였고, 기존의 DCPA와 TCPA를 이용한 퍼지 추론 방법보다 효율적으로 충돌 위험도를 추정할 수 있다.

• 대한임베디드공학회논문지
• /
• 제19권2호
• /
• pp.91-99
• /
• 2024

Collision avoidance system is important to improve the robustness and functional safety of autonomous vehicles. This paper proposes an object-level distance estimation method to develop a collision avoidance system, and it is applied to golfcarts utilized in country club environments. To improve the detection accuracy, we continually trained an object detection model based on pseudo labels generated by a pre-trained detector. Moreover, we propose object-aware depth estimation (OADE) method which trains a depth model focusing on object regions. In the OADE algorithm, we generated dense depth information for object regions by utilizing detection results and sparse LiDAR points, and it is referred to as object-aware LiDAR projection (OALP). By using the OALP maps, a depth estimation model was trained by backpropagating more gradients of the loss on object regions. Experiments were conducted on our custom dataset, which was collected for the travel distance of 22 km on 54 holes in three country clubs under various weather conditions. The precision and recall rate were respectively improved from 70.5% and 49.1% to 95.3% and 92.1% after the continual learning with pseudo labels. Moreover, the OADE algorithm reduces the absolute relative error from 4.76% to 4.27% for estimating distances to obstacles.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2010년도 추계학술대회
• /
• pp.180-181
• /
• 2010

선박에서 항해사의 안전운항을 효과적으로 지원하고 충돌사고를 방지하기 위해 다중선박의 충돌 위험도를 추정하는 알고리즘을 개발하였다. 선박 충돌위험도 추정 알고리즘은 선박들의 항행정보로서, AIS 정보를 사용하고, 퍼지 이론을 이용하여 충돌위험도를 계산한다. 지난 연구에서는, 고안된 알고리즘의 성능을 검증하기 위해 울산항 해상교통관제(VTS) 센터로부터 수집된, 실제 울산항에서 운행된 선박들의 AIS데이터를 기반으로 검증 시뮬레이션을 수행한 바 있다. 본 논문에서는 선박 충돌위험도 추정 알고리즘을 좀 더 정밀하게 검증하기 위해, 실제 해상충돌사고 데이터에 적용해 보고자 하였다. 이를 위해, 2009년 부산항에서 발생한 석유제품 운반선과 화물선간의 충돌사고에 대한 AIS 데이터를 수집하였고, 이를 이용하여 선박운항 시뮬레이터 기반, 재생 시뮬레이션을 수행하였다. 본 논문에서는 선박 충돌위험도 추정 알고리즘의 특징과, 실제 선박 충돌사고의 AIS 데이터에 대한 재생 시뮬레이션 결과에 대해 소개하였다.

• 한국항해항만학회지
• /
• 제34권10호
• /
• pp.727-733
• /
• 2010

해양사고에서 선박간의 충돌사고가 많은 부분을 차지하고 있으며, 상당수가 인적오류에 의해 발생되고 있다. 본 논문에서는 선박에서 항해사의 안전운항을 효과적으로 지원하고 충돌사고를 방지하기 위해 다중선박의 충돌 위험도를 추정하는 알고리즘을 개발하였다. 선박 충돌위험도 추정 알고리즘은 선박들의 항행정보로서 AIS 정보를 사용하고 퍼지 이론을 이용하여 충돌위험도를 계산한다. 지난 연구에서는 고안된 알고리즘의 성능을 검증하기 위해 울산항 해상교통관제(VTS) 센터로부터 실제 울산항에서 운행된 선박들의 AIS데이터를 수집하였고, 이를 기반으로 검증 시뮬레이션을 수행한 바 있다. 본 논문에서는 선박 충돌위험도 추정 알고리즘을 좀더 정밀하게 검증하기 위해 실제 해상충돌사고 데이터에 적용해 보고자 하였다. 이를 위해, 2009년 부산항에서 발생한 석유제품 운반선과 화물선간의 충돌사고에 대한 AIS 데이터를 수집하였고 이를 이용하여 선박운항 시뮬레이터 기반 재생 시뮬레이션을 수행하였다. 시뮬레이션 결과, 실제 사고 상황에 적용할 경우 충돌 사고가 일어나기 전에 충돌 위험을 표시하여 충돌사고를 경고할 수 있음을 확인할 수 있었다.

• 한국콘크리트학회:학술대회논문집
• /
• 한국콘크리트학회 2004년도 춘계 학술발표회 제16권1호
• /
• pp.416-419
• /
• 2004

The kinetic energy during ship collision with bridge piers is released as the permanent deformations of structure and friction between the impact surfaces. So the ship collision energy is estimated from the equations of motions for ship-pier collisions which include the influence of the surrounding water, different impact angles and impact locations. The normal impact energy and tangent impact energy at a collision location and angle can be transformed into the normal impact force and friction force acting on the structure. Also the kinetic energy after collisions is calculated from the linear and angular impulse of ship collisions. The collision energy absorption system such as the protective structures for bridges is designed by evaluating the damage portions of ship and structure during the ship-structure collisions varying from the soft impact to hard impact and then the estimation of it will be suited for the design of protective measures.

• 자동차안전학회지
• /
• 제5권2호
• /
• pp.17-23
• /
• 2013

This paper describes an algorithm for Advanced Emergency Braking(AEB) with tire-road friction coefficient estimation. The AEB is a system to avoid a collision or mitigate a collision impact by decelerating the car automatically when forward collision is imminent. Typical AEB system is operated by Time-to-collision(TTC), which considers only relative velocity and clearance from control vehicle to preceding vehicle. AEB operation by TTC has a limit that tire-road friction coefficient is not considered. In this paper, Tire-road friction coefficient is also considered to achieve more safe operation of AEB. Interacting Multiple Model method(IMM) is used for Tire-road friction coefficient estimation. The AEB algorithm consists of friction coefficient estimator and upper level controller and lower level controller. The numerical simulation has been conducted to demonstrate the control performance of the proposed AEB algorithm. The simulation study has been conducted with a closed-loop driver-controller-vehicle system using using MATLAB-Simulink software and CarSim Vehicle model.

• 한국정보기술응용학회:학술대회논문집
• /
• 한국정보기술응용학회 2005년도 6th 2005 International Conference on Computers, Communications and System
• /
• pp.203-206
• /
• 2005

Recently, the RFID(Radio Frequency Identification) technology has gained significant attention. One of the performance issues in RFID systems is to resolve the tag collision among responses from RFID tags. In this paper, we proposed a new scheme for estimation of the number of tags in the reader filed. The scheme is used by anti-collision algorithm to identify multiple tags efficiently. And we also present the simulation result that shows the proposed scheme to estimate tags efficiently and also to improve the systems efficiency.