• Agricultural and Biosystems Engineering
• /
• v.2 no.2
• /
• pp.69-74
• /
• 2001

Detection of operation lines in farm works, object recognition and obstacle avoidance are essential pre-requisite technologies for unmanned agricultural machinery. A CCD camera, which has been largely used for these functions, is expensive and has difficulty in real-time signal processing. In this study, a laser range sensor was selected as the guiding vision for unmanned agricultural machinery such as a tractor. To achieve this capability, algorithms for distance measurement, signal filtering, object recognition, and obstacle avoidance were developed. Computer simulations were carried out to evaluate performance of the algorithms. Experiments were also conducted with various materials and shapes, Laser beam lost its intensity for poor reflective materials, resulting in less range value than actual, so a compensation technique was considered to be necessary. Object detection system was fabricated on an agricultural tractor and the performance was evaluated. As test result for obstacle detection and avoidance in field, to detect and avoid obstacle for path finding with guiding system for unmanned agricultural machinery was enable.

• Journal of Advanced Navigation Technology
• /
• v.10 no.2
• /
• pp.145-151
• /
• 2006

In this paper, we propose a novel navigation method combined Nearness Diagram, Limit-Cycle method and the Vector Field Method for avoidance of unexpected obstacles in the dynamic environment. The Limit-Cycle method is used to obstacle avoidance in front of the robot and the Vector Field Method is used to obstacle avoidance in the side of robot. And the Nearness Diagram Navigation is used to obstacle avoidance in the nearness area of the robot. The performance of the proposed method is demonstrate by simulations.

• Journal of the Korean Institute of Telematics and Electronics T
• /
• v.35T no.3
• /
• pp.67-75
• /
• 1998

This paper presents a new robot motion planning method for moving obstacle avoidance. To consider the mobility of a moving obstacle, we define virtual distance function(VDF) between the robot and the obstacle. At each sampling time, we use the VDF to construct an artificial potential, considering the motion of obstacles. The robot moves according to the repulsive and attractive force vector induced by the artificial potential function. The proposed algorithm can be driven the robot to avoid moving obstacles in real time. Some simulation studies show the effectiveness of the proposed method.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1995.10a
• /
• pp.327-330
• /
• 1995

If a mobile robot is used in a real situation, robot must face a moving obstacles. In that case, the collision avoidance algorithm for moving obstacle is a indispensible element in mobile robot control. We csrried out a research to find and evaluate the advanced algorithm for mobile robot. At first we generate the continous path for mobi;e robot. Then by creating a curved path for avoidance, the mobile robot can change its path smoothly. Smoothed path made the robot adapt more effectively to the changing of path. Under time-varying condition, computer simulation was performed to show the validation of proposed algorithm.

• International Journal of Aeronautical and Space Sciences
• /
• v.6 no.2
• /
• pp.97-109
• /
• 2005

In this paper, the critical requirement for obstacle awareness and avoidance is assessed with the compliance of the equivalent level of safety regulation, and then the collision avoidance sensor system is presented with the key design parameters for the requirement of the smart unmanned aerial vehicle in low-altitude flight. Based on the assessment of various sensors, small-sized radar sensor is selected for the suitable candidate due to the real-time range and range-rate acquisition capability of the stationary and moving aircraft even under all-weather environments. Through the performance analysis for the system requirement, the conceptual design result of radar sensor model is proposed with the range detection probability and collision avoidance mode is established based on the time-to-collision, which is analyzed by collision scenario.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.7-12
• /
• 1992

This paper proposes a sensor-based path planning method which utilizes fuzzy logic and neural network for obstacle avoidance of a mobile robot in uncertain environments. In order to acquire the information about the environment around the mobile robot, the ultrasonic sensors mounted on the front of mobile robot are used. The neural network, whose inputs are preprocessed by ultrasonic sensor readings, informs the mobile robot of the situation of environment in which mobile robot is at the present instant. Then, according to the situation class, the fuzzy rules are fired to make a decision on the mobile robot action. In addition, this method can be implemented real time since the number of fuzzy rules used to avoid the obstacle is small. Fuzzy rules are constructed based on the human reasoning and tuned by iterative simulations. The effective of the proposed avoidance method is verified by a series of simulations.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.7
• /
• pp.1302-1308
• /
• 2007

In this paper, a real-time collision avoidance algorithm is proposed for experimental Autonomous Vehicle(XAV). To ensure real-time implementation, a virtual potential field is calculated in one dimensional space. The attractive force is generated by the steering command either transmitted in the remote control station or calculated in the Autonomous Navigation System(ANS) of the XAV. The repulsive force is generated by obstacle information obtained from Laser Range Finder(LRF) mounted on the XAV. Using these attractive and repulsive forces, modified steering, velocity and emergency stop commands are created to avoid obstacles and follow a planned path. The suggested algorithm is inserted as one component in the XAV system. Through various real experiments and technical demonstration using the XAV, the usefulness and practicality of the proposed algorithm are verified.

• Journal of the Korean Society for Precision Engineering
• /
• v.27 no.2
• /
• pp.60-69
• /
• 2010

Lookup-Table (LUT) based fuzzy controller for obstacle avoidance enhances operations faster in multiple obstacles environment. An LUT based fuzzy controller with Positive/Negative (P/N) fuzzy rule base consisting of 18 rules was introduced in our paper$^1$ and this paper shows a 50-rule P/N fuzzy controller for enhancing performance in obstacle avoidance. As a rule, the more rules are necessary, the more buffers are required. This paper suggests LUT sharing method in order to reduce LUT buffer size without significant degradation of performance. The LUT sharing method makes buffer size independent of the whole fuzzy system's complexity. Simulation using MSRDS(MicroSoft Robotics Developer Studio) evaluates the proposed method, and in order to investigate its performance, experiments are carried out to Pioneer P3-DX in the LabVIEW environment. The simulation and experiments show little difference between the fully valued LUT-based method and the LUT sharing method in operation times. On the other hand, LUT sharing method reduced its buffer size by about 95% of full valued LUT-based design.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.2
• /
• pp.53-63
• /
• 2018

A novel real-time path planning algorithm suitable for implementation on a small mobile robot is introduced. The algorithm can be used as the basis for mapping unknown or partially known environments and is tested in a specially developed simulation environment in Matlab(R). Simulations results are presented demonstrating that the algorithm can readily be implemented to allow a small robot to navigate in various unknown and partially known environments. The main characteristics of the algorithm include simplicity, ease of implementation, speed, and efficiency, thereby being especially suitable for small robots. Furthermore, for partially known environments, another algorithm is proposed to predefine an optimal path taking into account information provided regarding the environment.

• The Journal of Korea Robotics Society
• /
• v.12 no.3
• /
• pp.297-305
• /
• 2017

As the development of autonomous vehicles becomes realistic, many automobile manufacturers and components producers aim to develop 'completely autonomous driving'. ADAS (Advanced Driver Assistance Systems) which has been applied in automobile recently, supports the driver in controlling lane maintenance, speed and direction in a single lane based on limited road environment. Although technologies of obstacles avoidance on the obstacle environment have been developed, they concentrates on simple obstacle avoidances, not considering the control of the actual vehicle in the real situation which makes drivers feel unsafe from the sudden change of the wheel and the speed of the vehicle. In order to develop the 'completely autonomous driving' automobile which perceives the surrounding environment by itself and operates, ability of the vehicle should be enhanced in a way human driver does. In this sense, this paper intends to establish a strategy with which autonomous vehicles behave human-friendly based on vehicle dynamics through the reinforcement learning that is based on Q-learning, a type of machine learning. The obstacle avoidance reinforcement learning proceeded in 5 simulations. The reward rule has been set in the experiment so that the car can learn by itself with recurring events, allowing the experiment to have the similar environment to the one when humans drive. Driving Simulator has been used to verify results of the reinforcement learning. The ultimate goal of this study is to enable autonomous vehicles avoid obstacles in a human-friendly way when obstacles appear in their sight, using controlling methods that have previously been learned in various conditions through the reinforcement learning.