Collision avoidance using neural networks
01 Nov 2017-Vol. 263, Iss: 5, pp 052041
TL;DR: A model (robot) is developed to assist drivers for a smooth travel without accidents that reacts to the real time obstacles on the four critical sides of the vehicle and takes necessary action.
Abstract: Now a days, accidents on roads are caused due to the negligence of drivers and pedestrians or due to unexpected obstacles that come into the vehicle's path. In this paper, a model (robot) is developed to assist drivers for a smooth travel without accidents. It reacts to the real time obstacles on the four critical sides of the vehicle and takes necessary action. The sensor used for detecting the obstacle was an IR proximity sensor. A single layer perceptron neural network is used to train and test all possible combinations of sensors result by using Matlab (offline). A microcontroller (ARM Cortex-M3 LPC1768) is used to control the vehicle through the output data which is received from Matlab via serial communication. Hence, the vehicle becomes capable of reacting to any combination of real time obstacles.
Citations
More filters
Dissertation•
01 Oct 2019
TL;DR: A novel survey of the state-of-the-art geometric approaches with respect to performance, scalability and tolerance to uncertainty is conducted, and the interval avoidance (IA) method is proposed, to incorporate trajectory uncertainty in the geometric derivation of escape trajectories.
Abstract: Coordination is fundamental component of autonomy when a system is defined by multiple mobile agents. For unmanned aerial systems (UAS), challenges originate from their low-level systems, such as their flight dynamics, which are often complex. The thesis begins by examining these low-level dynamics in an analysis of several well known UAS using a novel symbolic component-based framework. It is shown how this approach is used effectively to define key model and performance properties necessary of UAS trajectory control. This is demonstrated initially under the context of linear quadratic regulation (LQR) and model predictive control (MPC) of a quadcopter.
The symbolic framework is later extended in the proposal of a novel UAS platform, referred to as the ``Polycopter" for its morphing nature. This dual-tilt axis system has unique authority over is thrust vector, in addition to an ability to actively augment its stability and aerodynamic characteristics. This presents several opportunities in exploitative control design.
With an approach to low-level UAS modelling and control proposed, the focus of the thesis shifts to investigate the challenges associated with local trajectory generation for the purpose of multi-agent collision avoidance. This begins with a novel survey of the state-of-the-art geometric approaches with respect to performance, scalability and tolerance to uncertainty. From this survey, the interval avoidance (IA) method is proposed, to incorporate trajectory uncertainty in the geometric derivation of escape trajectories. The method is shown to be more effective in ensuring safe separation in several of the presented conditions, however performance is shown to deteriorate in denser conflicts.
Finally, it is shown how by re-framing the IA problem, three dimensional (3D) collision avoidance is achieved. The novel 3D IA method is shown to out perform the original method in three conflict cases by maintaining separation under the effects of uncertainty and in scenarios with multiple obstacles. The performance, scalability and uncertainty tolerance of each presented method is then examined in a set of scenarios resembling typical coordinated UAS operations in an exhaustive Monte-Carlo analysis.
9Â citations
Cites background from "Collision avoidance using neural ne..."
...Direct applications of neural networks (NN) to collision avoidance in autonomous systems can be found in [97, 171, 235, 238]....
[...]
26 Aug 2020
TL;DR: The application of artificial neural network algorithms that get input in the form of signals that refer to the output signal from the ultrasonic sensors, which have already assembled into a multi-sensor, which is 8 (eight) ultrasonic sensor positioned around the vehicle, which will support the autonomous vehicle system.
Abstract: Autonomous vehicles are vehicles that run automatically without a driver. Therefore, the vehicle requires sensors to detect surrounding objects to avoid collisions with other objects or vehicles. A variety of sensors can be used to support the system. One of the sensors used is an ultrasonic sensor that has the reliability and robustness to various light conditions and radio waves, especially when compared to camera sensors, radar and lidar. The recent implementation of ultrasonic sensors in vehicles is limited as a parking guide, so it needs to be developed for further functions, considering that ultrasonic sonar technology has advanced with even greater detection and long distances range. Hence, as a continuation of previous research in ultrasonic sensor characteristics, this paper carries out the application of artificial neural network algorithms that get input in the form of signals that refer to the output signal from the ultrasonic sensors, which have already assembled into a multi-sensor, which is 8 (eight) ultrasonic sensors positioned around the vehicle, two sensors in the front, two sensors in the rear and four sensors in the right and left side of vehicle. The sensors and algorithms will support the autonomous vehicle system, where if the sensors detect the obstructive objects, the system will provide an output in the form of a decision to make the braking order, soft braking, turning left, turning right, or staying run straight when the front sensors do not detect a barrier object. This is done in anticipation of an accident and avoid a collision. Each condition and decision will be determined by which sensor detects the barrier object. Input and output will be simulated using the tool of artificial neural network algorithms, so as to get the most optimal weight and low error rate.
3Â citations
TL;DR: Ahmad Zuber Ahmad Zainuddin, Wahidah Mansor, Khuan Yoot Lee, Zulkifli Mahmoodin Microwave Research Institute, Universiti Teknologi MARA, Shah Alam, Malaysia School of Electrical Engineering, College of Engineering, University of Malaysia, Malaysia Computational Intelligence Detection RG, Health and Wellness ReNEUNEU, Universiteng N.
Abstract: Ahmad Zuber Ahmad Zainuddin, Wahidah Mansor, Khuan Yoot Lee, Zulkifli Mahmoodin Microwave Research Institute, Universiti Teknologi MARA, Shah Alam, Malaysia School of Electrical Engineering, College of Engineering, Universiti Teknologi MARA, Shah Alam, Malaysia Computational Intelligence Detection RG, Health and Wellness ReNEU, Universiti Teknologi MARA, Shah Alam, Malaysia Medical Engineering Technology Section, Universiti Kuala Lumpur British Malaysian Institute, Malaysia
References
More filters
Book•
10 Jan 2014
TL;DR: In this paper, the potential advantages of autonomous vehicles, including safety, mobility and fuel consumption, and disadvantages, including travel demand, are discussed, and guidance for policymakers and risks to be considered as policies are formulated.
Abstract: Autonomous vehicles may change the transportation landscape and policymakers must find the best course to maximize social benefits while minimizing drawbacks. This report weighs the potential advantages, including safety, mobility and fuel consumption, and disadvantages, including travel demand. Current legislation in Nevada, Florida, California, and Washington, DC, and in other states, and the standards and regulations that apply to such vehicles, are outlined. The report provides a brief history of autonomous vehicles and a summary of the role of telematics and communications, and discusses liability for drivers, insurers and manufacturers. It contains guidance for policymakers and risks to be considered as policies are formulated.
769Â citations
01 Jan 2015
TL;DR: In this article, the authors explore the impacts that autonomous vehicles are likely to have on travel demands and transportation planning and explore how they will affect planning decisions such as optimal road, parking and public transit supply.
Abstract: This paper explores the impacts that autonomous (also called self-driving, driverless or robotic) vehicles are likely to have on travel demands and transportation planning. It discusses autonomous vehicle benefits and costs, predicts their likely development and implementation based on experience with previous vehicle technologies, and explores how they will affect planning decisions such as optimal road, parking and public transit supply. The analysis indicates that some benefits, such as independent mobility for affluent non-drivers, may begin in the 2020s or 2030s, but most impacts, including reduced traffic and parking congestion (and therefore road and parking facility supply requirements), independent mobility for low-income people (and therefore reduced need to subsidize transit), increased safety, energy conservation and pollution reductions, will only be significant when autonomous vehicles become common and affordable, probably in the 2040s to 2060s, and some benefits may require prohibiting human-driven vehicles on certain roadways, which could take longer.
764Â citations
TL;DR: The paper briefly summarizes the approaches that different teams used in the 2007 DARPA Urban Challenge, with the goal of describing some of the challenges that the teams faced in driving in urban environments.
Abstract: The development of autonomous vehicles for urban driving has seen rapid progress in the past 30 years. This paper provides a summary of the current state of the art in autonomous driving in urban environments, based primarily on the experiences of the authors in the 2007 DARPA Urban Challenge (DUC). The paper briefly summarizes the approaches that different teams used in the DUC, with the goal of describing some of the challenges that the teams faced in driving in urban environments. The paper also highlights the long-term research challenges that must be overcome in order to enable autonomous driving and points to opportunities for new technologies to be applied in improving vehicle safety, exploiting intelligent road infrastructure and enabling robotic vehicles operating in human environments.
388Â citations
Journal Article•
TL;DR: The present study provides the magnitude and various dimensions of road accident in India and the analysis on road accidents will help to create awareness, guidelines and assist in informed decision making on road safety.
Abstract: Road accidents are a human tragedy. They involve high human suffering and monetary costs in terms of untimely deaths, injuries and loss of potential income. Although we have undertaken many initiatives and are implementing various road safety improvement program the overall situation as revealed by data is far from satisfactory. During the calendar year 2010, there were close to 5 lakh road accidents in India, which resulted in more than 1.3 lakh persons. These numbers translate intone road accident every minute, and one road accident death every 4 minutes. Unfortunately, more than half the victims are in the economically active age group of 25-65 years. The loss of the main bread winner can be catastrophic. Road traffic accidents are amenable to remedial action. Many a countries have curbed the menace of road accidents by adopting a multipronged approach to road safety that encompasses broad range of measures, such as, traffic management, design and quality of road infrastructure, application of intelligent transport system, safer vehicles, law enforcement, effective and quick accident response and care etc. The Government alone cannot tackle road safety problems. There is a need for active involvement of all stake- holders to promote policy reform and implementation of road safety measures. Addressing road safety is comprehensive manner underscores the need to involve multiple agencies and sectors like health, transport and police. The present study provides the magnitude and various dimensions of road accident in India. The analysis on road accidents in this study will help to create awareness, guidelines and assist in informed decision making on road safety. Language: en
26Â citations
Proceedings Article•
04 Jun 1996TL;DR: A neural netwok learned by genetic algorithm is introduced to solve conflicts between two aircrafts and results are very promising.
Abstract: As Air Traffic keeps increasing, many research programs focus on collision avoidance techniques. In this paper, a neural netwok learned by genetic algorithm is introduced to solve conflicts between two aircrafts. The learned NN is then tested on different conflicts and compared to the optimal solution. Results are very promising.
12Â citations