Automatic Collision Avoidance System Research Articles

The technique for approximation of three-dimensional surfaces in order to synthesize algorithms for preventing aircraft collisions with obstacles

The work marks the beginning of the practical application of algorithms for making the aircraft recovery maneuver from three-dimensional constraint surfaces, which are a combination of terrain and artificial obstacles. An analysis of events leading to aviation accidents was carried out, and a comparison of on-board systems for preliminary notification of aircraft crews about collisions with natural or artificial obstacles was made. It is shown that such systems are insufficient due to their passive-recommendatory nature of issuing warnings. A question has been raised about the need to implement an active automatic collision avoidance system with spatial obstacles. In order to apply existing algorithms for the aircraft recovery maneuver from spatial constraint surfaces, a technique has been developed for approximating three-dimensional surfaces (obstacles) specified on a digital terrain map in the form of discrete height readings with a certain step on a coordinate grid. A paraboloid of revolution was chosen as a continuous 2nd order surface approximating the obstacle, and its characteristic parameters were determined. To determine the characteristic parameters of the paraboloid, an algorithm for determining the intersection of a three-dimensional surface and a plane, based on the principle of determining the intersection of triangles in space, as well as a method for selecting the inflection point of the terrain, based on determining the value of the terrain height gradient, are proposed for use. The construction of an approximating paraboloid using the example of a natural obstacle in the form of a mountain range is given. When synthesizing algorithms for preventing collisions of aircraft with obstacles, the need to take into account not only the parameters of the constraint surfaces and dynamic characteristics of aircraft, but also the accuracy characteristics of data sources about their position is noted. Promising application areas of the developed methodology are shown.

Human-like route planning for automatic collision avoidance using generative adversarial imitation learning

Automatic collision avoidance systems are expected to eliminate maritime accidents caused by human error. Recent studies have shown that ships can prevent collisions in complex situations by using deep reinforcement learning (DRL). However, rewards must be designed to implement DRL, which can be challenging for ambiguous tasks, such as following the Convention on the International Regulations for Preventing Collisions at Sea (COLREGs). While some studies have used inverse reinforcement learning (IRL) to derive an appropriate reward for limited problems with small, discrete state spaces, this study introduced generative adversarial imitation learning (GAIL) and proposed a route planning algorithm that addresses the above limitations. We applied DRL to generate sample collision avoidance trajectories and demonstrate that the imitative route planner based on GAIL can reproduce these trajectories. Also, expert trajectories were collected in simulation experiments involving a well-experienced captain, and we attempted to generate collision avoidance routes that mimic human expert performance. By applying a convolutional neural network and other techniques to improve the planner's imitation capability, we established a sophisticated route planner that is consistent with captain's risk assessments and tolerant of noise.

Failure behavior of multi-state system considering functional and physical isolation effects

The multi-state system (MSS) is a system that may exhibit multiple states or performance levels. Most existing studies assumed that the transition probabilities from states to states are known. However, in practical engineering, the complex stress conditions lead to great difficulty of collecting the statistical data of state transitions. In this paper, based on physics-of-failure theory, we consider different levels of damages caused by failure mechanisms (FMs) are the main reasons to components’ multiple states. Besides, the physical isolation (PI) effect on degradations of FMs is also studied, which is neglected in the existing studies about the MSS with functional dependence groups. Decision-diagram based methods are used for modeling the failure behavior of the MSS. An automatic collision avoidance system is analyzed for illustrating the proposed modeling and analyzing methods. The results show that comparing to the results without the consideration of PI effect, the probabilities of different states with PI effect of multi-state components and system may decrease or increase, which depends on the actual PI effects to the stress conditions.

Implications of mmWave Radiation on Human Health: State of the Art Threshold Levels

millimeter (mmWave) frequencies are covering from 30GHz to 300GHz in the electromagnetic spectrum and their uses in various applications like next-generation wireless communication systems (massive 5G telecommunications network), medical devices, airport security and automatic collision avoidance systems are growing vastly in the near future. Therefore, it is important to study the effects of mmWave radiation (non-ionization radiation) on biological systems and biophysical mechanisms. This paper focus on thorough review of nascent literature about current understandings of biological effects and epidemiological studies due to mmWave radiation in human beings. It presents latest guidelines with quantitative electromagnetic field thresholds by considering the realistic exposure scenarios of “general public” and “occupational” who undergo through wireless communication sources in their daily life. It also gives necessary safety measures to be taken while using the emerging mmWave technologies for future generation wireless communication networks.

Automatic collision avoidance system with many targets, including maneuvering ones

The article considers the issues of automatic collision avoidance with many targets, including maneuvering ones, which is especially important when navigating a vessel in narrownesses. Was made a brief review of literature devoted to the problem of collision avoidance and conclusion was drown on the relevance of developing such systems. There were developed mathematical and algorithmic support of the combined collision avoidance with many targets, including maneuvering ones. Based on the synthesized algorithms, there was developed software in the MATLAB environment for a simulator of an onboard controller of a ship control system, the operability and efficiency of mathematical, algorithmic and software was tested in a closed circuit with a Navi Trainer 5000 navigation simulator for various types of ships, navigation areas and weather conditions. The experiments confirmed the efficiency of the proposed method and algorithms and allow to recommend them for practical use in the development of modules for automatic collision avoidance with many targets, including maneuvering ones, an onboard controller of the ship’s control system.

Reliability Evaluation and Failure Behavior Modeling of IMS Considering Functional and Physical Isolation Effects

Due to the integration of complex structure and artificial intelligence, the failure behavior of intelligent mechatronic system (IMS) is characterized by functional and physical dependences. In this article, the functional isolation and physical isolation (PI) effects caused by these dependences are studied. We not only analyze the PI effect affecting the failure behaviors of components through failure mechanisms, but also propose a hierarchical method based on binary decision diagram for modeling system failure behavior affected by the PI and functional isolation effect. In the case study, an automatic collision avoidance system as an example of IMS is evaluated for the analysis of failure behavior, which is helpful for more realistic reliability assessment and design optimization.

Study on Automatic Collision Avoidance System and Method for Evaluating Collision Avoidance Manoeuvring Results

The authors have two goals that want to achieve. One is development of an automatic collision avoidance system for unmanned autonomous ships. And the other one is development of evaluation method of automatic collision avoidance manoeuvring results. In this paper, the outline of the developed automatic collision avoidance system and the result of the verification test on the actual ship navigating in congested waters are introduced. And also a discussion of the relationship between COLREGs and the desirable automatic collision avoidance system and methods for quantitatively assessing the desirable automatic collision avoidance system are introduced. In addition, quantitative comparison by the proposed evaluation system between the manoeuvring results of the automatic collision avoidance system and the veteran captain’s manoeuvring results are introduced.

Assisting the Improvement of a Military Safety System: An Application of Rapid Assessment Procedures to the Automatic Ground Collision Avoidance System

This article describes an iterative application of Rapid Assessment Procedures (RAP) to study human-machine interaction issues with a recently implemented, highly automatic system on the F-16 fighter aircraft known as the Automatic Ground Collision Avoidance System (Auto-GCAS). Auto-GCAS is a sophisticated technology that has the ability to detect an imminent ground collision threat and automatically execute a last-moment maneuver to avoid a crash. We employed RAP methods at multiple United States Air Force and Air National Guard sites in the United States and abroad. Over a three-year period, we conducted semi-structured interviews with 402 F-16 pilots experienced with the system. The method we employed, termed here iterative RAP, is reviewed in detail and evaluated in the framework of Utarini, Winkvist, and Pelto's (2001) eleven criteria for quality RAP studies. Results from this study include assisting in correcting system misunderstandings and anomalies, improving information flow about Auto-GCAS, and contributing towards the perception of military safety systems as valuable. This article concludes by (1) discussing positive effects that iterative RAP can contribute to the defense community and (2) arguing our method's utility towards the study of complex bureaucracies and multi-sited research, particularly following the introduction of new technology or policy.

AN AUTOMATIC COLLISION AVOIDANCE AND ROUTE GENERATING ALGORITHM FOR SHIPS BASED ON FIELD MODEL

This study used a velocity potential field approach to develop an automatic collision avoidance system and a route-generating algorithm for merchant ships. The system includes course-changing and track-keeping modes. The system based on the velocity potential of source/vortex and dipole flow theory in fluid dynamics. The course-changing mode creates a source/vortex flow vector to guide the ship away from various obstacles. The track-keeping mode generates a dipole flow field to steer the ship back to the desired course. An algorithm for collision avoidance maneuvers is implemented by using the real-time data of DCPA, TCPA, and bearing angle based on the results of maneuver simulation. Collision prevention regulations and international navigational rules are incorporated into the algorithm. The velocity potential field method is straightforward and very simple to implement; it avoids the problem of deadlock that hampers artificial potential methods and is suitable for the real-time path planning of complex static obstacles or ship encounter situations. The method has been applied to some typical test cases, including head-on, over-take, and crossing encounter situations with ships. The results prove that the proposed method is useful in finding safe paths for ships in encounter situations.

Automatic collision avoidance of multiple ships based on deep Q-learning

As the number of ships for marine transportation increases with the globalisation of the world economy, waterways are becoming more congested than before. This situation will raise the risk of collision of the ships; hence, an automatic collision avoidance system needs to be developed. In this paper, a novel approach based on deep reinforcement learning (DRL) is proposed for automatic collision avoidance of multiple ships particularly in restricted waters. A training method and algorithms for collision avoidance of ships, incorporating ship manoeuvrability, human experience and navigation rules, are presented in detail. The proposed approach is investigated not only by numerical simulations but also by model experiments using three self-propelled ships. Through the systematic numerical and experimental validation, it is demonstrated the developed approach based on the DRL has great possibility for realising automatic collision avoidance of ships in highly complicated navigational situations.

심층 결정론적 정책 경사법을 이용한 선박 충돌 회피 경로 결정

The stability, reliability and efficiency of a smart ship are important issues as the interest in an autonomous ship has recently been high. An automatic collision avoidance system is an essential function of an autonomous ship. This system detects the possibility of collision and automatically takes avoidance actions in consideration of economy and safety. In order to construct an automatic collision avoidance system using reinforcement learning, in this work, the sequential decision problem of ship collision is mathematically formulated through a Markov Decision Process (MDP). A reinforcement learning environment is constructed based on the ship maneuvering equations, and then the three key components (state, action, and reward) of MDP are defined. The state uses parameters of the relationship between own-ship and target-ship, the action is the vertical distance away from the target course, and the reward is defined as a function considering safety and economics. In order to solve the sequential decision problem, the Deep Deterministic Policy Gradient (DDPG) algorithm which can express continuous action space and search an optimal action policy is utilized. The collision avoidance system is then tested assuming the 90°intersection encounter situation and yields a satisfactory result.

Simulation of Automatic Pedestrian Collision Avoidance System in Autonomous Driving of Personal Mobility Vehicle

Simulation of Automatic Pedestrian Collision Avoidance System in Autonomous Driving of Personal Mobility Vehicle

Development of Automatic Collision Avoidance System and Quantitative Evaluation of the Manoeuvring Results

Development of Automatic Collision Avoidance System and Quantitative Evaluation of the Manoeuvring Results

Design of an Automatic Collision Avoidance System for Ships Based on Wireless Sensor Network

The system designed is a ship collision avoidance system based on consistent use of satellite positioning technology, spread spectrum communication technology and a wireless sensor network. The system design includes: an information collecting terminal, a data processing terminal and a mobile data terminal as the three main parts. CC2530 is selected as the master chip for the information collecting terminal, and the GPS module with NEO-6M UBLOX satellite positioning function is used to obtain the latitude, longitude, heading and other information. The AS62-T30 wireless communication module is used to realize the data interaction between ships, and a 0.96-inch OLED display module is used to show the current location of the ship, thus realizing the GPS positioning data receiving, data analysis, information display, data integration and transmission, and other functions. In terms of the software in the data processing terminal, QT5 is selected as the development environment, and QtSql as the database to process and store the data packet sent by the information collecting terminal. The system has many functions including real-time data analysis and alarm, real-time location annotation, track query, route planning and weather forecasting, etc.

Study on Automatic Collision Avoidance System in Autonomous Vehicle

Study on Automatic Collision Avoidance System in Autonomous Vehicle

Automatic collision avoidance system for four-wheel independent drive vehicle by state-dependent Riccati equation

Automatic collision avoidance system for four-wheel independent drive vehicle by state-dependent Riccati equation

A review of essential technologies for collision avoidance assistance systems

Compared with automatic collision avoidance systems, collision avoidance assistance systems have attracted more research interest because they help avoid collisions in near-accident situations. However, ensuring the robustness and reliability of collision avoidance assistances is difficult because of problems in reliable environment recognition, accurate collision avoidance decision making, and driver acceptance. This article reviews and analyzes three essential technologies for collision avoidance assistance systems in intelligent environment-friendly vehicles. The architecture of the human-in-loop system is summarized, and available environment recognition methods used in collision avoidance assistances are compared. Moreover, the characteristics of path planning and collision avoidance decision making applied in emergency collision avoidance assistance systems are reviewed, and the advantages of applying haptic share control and yaw moment active intervention during path tracking are analyzed. The emer...

Automatic Collision Avoidance System in Electric Vehicle with Independently Driven Four Wheels Considering Vehicle Velocity

Automatic Collision Avoidance System in Electric Vehicle with Independently Driven Four Wheels Considering Vehicle Velocity

An Automatic Collision Avoidance System Based on LiDAR

For vehicle-borne LiDAR, a mathematical model is built for the computation and reconstruction of laser point cloud with the scanning data, GPS data and IMU data. 3D point cloud of the road and the scenery on the both sides of the road is obtained. Then according to the trajectory of the vehicle, 3D roaming for the scenery on the both sides of the road is realized using OpenGL 3D engine technology. This technology provides a probably feasible way for anti-collision of vehicles and aircrafts when driven at night, in the heavy fog or flying between the mountains.

A user test of Automatic Navigational Intention Exchange Support System using an intelligent ship-handling simulator

Authors evaluated the feasibility of the “Automatic Navigational Intention Exchange Support System using AIS (ANIESS)”, which we have developed since 2007. ANIESS is a communication system to exchange navigational intensions (e.g. Port to Port passing) between encountered ships by using short-range communication means such as application specific message of automatic identification system (AIS). A series of simulator experiments for evaluation of the ANIESS were carried out participating five active seamen. The scenarios are composed of basic encounter situations. Each subject experienced the collision avoidance manoeuvre using the ANIESS and the automatic collision avoidance system (ACAS), which is installed to a simulator. Then, they assessed the collision avoidance manoeuvre using ANIESS subjectively regarding the way and the suitable time for the support of the ANIESS, comparing to their own usual operations. As the result, we found the appropriate way and time for the support. We also got some suggestions on the way and the time for starting collision avoidance manoeuvre to improve the ACAS. Finally, we confirmed the effectiveness in basic encounter condition from the subjective evaluation and also found some issues to improve the ANISS and ACAS.