Neural-network multiple models filter (NMM)-based position estimation system for autonomous vehicles

2013 ◽  
Vol 14 (2) ◽  
pp. 265-274 ◽  
Author(s):  
M. Gwak ◽  
K. Jo ◽  
M. Sunwoo
Keyword(s):  
Neural Network ◽  
Autonomous Vehicles ◽  
Position Estimation ◽  
Multiple Models ◽  
Estimation System

Optimal Configuration of Ultrasonic Sensors in a 3D Position Estimation System Using Genetic Algorithms

2001 ◽  
Author(s):  
Probir Kumar Ray ◽  
Nishant Unnikrishnan ◽  
Ajay Mahajan
Keyword(s):  
Autonomous Vehicles ◽  
Wave Theory ◽  
Position Estimation ◽  
Optimal Placement ◽  
Accurate Estimation ◽  
Delay Term ◽  
Electronic Circuitry ◽  
3D Space ◽  
The Difference ◽  
Estimation System

Abstract This paper provides a genetic algorithm based approach to calculate the optimal placement of receivers in a 3D position estimation system that uses the difference in the time-of-arrivals (TOA) of an ultrasonic wave from a transmitter to the different receivers fixed in 3D space. This is a different approach to traditional systems that use the actual time-of-flights (TOF) from the transmitter to the different receivers and triangulate the position of the transmitter. The new approach makes the system more accurate, makes the transmitter independent of the receivers and does not require the need of calculating the time delay term that is inherent in traditional systems due to delays caused by the electronic circuitry. This paper presents a thorough analysis of receiver configurations in the 2D and 3D system that lead to singularities, i.e. locations of receivers that lead to formulations that can not be solved due to a shortage of information. It provides guidelines of where not to place receivers, and further, presents a detailed analysis of locations that are optimal, i.e. locations that lead to the most accurate estimation of the transmitter positions. The results presented in this paper are not only applicable to ultrasonic systems, but all systems that use wave theory, e.g. infrared, laser, etc. This work finds applications in virtual reality cells, robotics, guidance of indoor autonomous vehicles and vibration analysis.

A position estimation system for mobile robots using a monocular image of a 3-D landmark

Robotica ◽  
1994 ◽  
Vol 12 (5) ◽  
pp. 431-441 ◽  
Author(s):  
Kyoung C. Koh ◽  
Jae S. Kim ◽  
Hyung S. Cho
Keyword(s):  
Neural Network ◽  
Mobile Robot ◽  
Estimation Algorithm ◽  
Position Estimation ◽  
Camera System ◽  
System A ◽  
The Neural Network ◽  
Monocular Image ◽  
Estimation System ◽  
Learning Data

SUMMARYThis paper presents an absolute position estimation system for a mobile robot moving on a flat surface. In this system, a 3-D landmark with four coplanar points and a non-coplanar point is utilized to improve the accuracy of position estimation and to guide the robot during navigation. Applying theoretical analysis, we investigate the image sensitivity of the proposed 3-D landmark compared with the conventional 2-D landmark. In the camera calibration stage of the experiments, we employ a neural network as a computational tool. The neural network is trained from a set of learning data collected at various points around the mark so that the extrinsic and intrinsic parameters of the camera system can be resolved. The overall estimation algorithm from the mark identification to the position determination is implemented in a 32-bit personal computer with an image digitizer and an arithmetic accelerator. To demonstrate the effectiveness of the proposed 3-D landmark and the neural network-based calibration scheme, a series of navigation experiments were performed on a wheeled mobile robot (LCAR) in an indoor environment. The results show the feasibility of the position estimation system applicable to mobile robot's real-time navigation.

Intelligent System Identification Using Neural Networks and Genetic Algorithms of an Ultrasonic Positioning System

2002 ◽  
Author(s):  
Nishant Unnikrishnan ◽  
Ajay Mahajan
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Genetic Algorithms ◽  
System Identification ◽  
Intelligent System ◽  
Position Estimation ◽  
Actual System ◽  
Neural Network Approach ◽  
Estimation System ◽  
The One

This paper presents an intelligent system identification methodology for the identification of a realistic model of an ultrasonic position estimation system that uses the difference in the time of arrivals of waves from a transmitter to various receivers. Even though a linearized formulation for the 3D system exists and is currently being used to estimate the position of the transmitter, its accuracy can still be improved further. A neural network approach is developed to train the system based on training sets obtained from the actual system, and it is proposed to use the final trained system to estimate the 3D position in real time. The weights of the neural network are obtained from an innovative procedure using genetic algorithms. Results for a simplified 1D system are presented as proof of concept. The performance of the identified 1D system using genetic algorithms is shown to be comparable to the one using the analytical model. Further, the identified system using genetic algorithms is also shown to be superior to the one using the traditional back propagation method for finding the weights for the neural networks. This work has significant applications in the identification of complex non-linear systems.

scholarly journals Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance

2021 ◽  
Vol 7 (4) ◽  
pp. 61
Author(s):  
David Urban ◽  
Alice Caplier
Keyword(s):  
Neural Network ◽  
Autonomous Vehicles ◽  
Depth Estimation ◽  
Video Camera ◽  
Obstacle Detection ◽  
Navigation Systems ◽  
Time To Collision ◽  
Static Data ◽  
Monocular Depth ◽  
Fully Connected

As difficult vision-based tasks like object detection and monocular depth estimation are making their way in real-time applications and as more light weighted solutions for autonomous vehicles navigation systems are emerging, obstacle detection and collision prediction are two very challenging tasks for small embedded devices like drones. We propose a novel light weighted and time-efficient vision-based solution to predict Time-to-Collision from a monocular video camera embedded in a smartglasses device as a module of a navigation system for visually impaired pedestrians. It consists of two modules: a static data extractor made of a convolutional neural network to predict the obstacle position and distance and a dynamic data extractor that stacks the obstacle data from multiple frames and predicts the Time-to-Collision with a simple fully connected neural network. This paper focuses on the Time-to-Collision network’s ability to adapt to new sceneries with different types of obstacles with supervised learning.

scholarly journals LVD-NMPC: A learning-based vision dynamics approach to nonlinear model predictive control for autonomous vehicles

2021 ◽  
Vol 18 (3) ◽  
pp. 172988142110195
Author(s):  
Sorin Grigorescu ◽  
Cosmin Ginerica ◽  
Mihai Zaha ◽  
Gigel Macesanu ◽  
Bogdan Trasnea
Keyword(s):  
Neural Network ◽  
Model Predictive Control ◽  
Predictive Control ◽  
Nonlinear Model ◽  
Autonomous Vehicles ◽  
Process Model ◽  
Network Performance ◽  
Vision System ◽  
Nonlinear Model Predictive Control ◽  
Scaled Model

In this article, we introduce a learning-based vision dynamics approach to nonlinear model predictive control (NMPC) for autonomous vehicles, coined learning-based vision dynamics (LVD) NMPC. LVD-NMPC uses an a-priori process model and a learned vision dynamics model used to calculate the dynamics of the driving scene, the controlled system’s desired state trajectory, and the weighting gains of the quadratic cost function optimized by a constrained predictive controller. The vision system is defined as a deep neural network designed to estimate the dynamics of the image scene. The input is based on historic sequences of sensory observations and vehicle states, integrated by an augmented memory component. Deep Q-learning is used to train the deep network, which once trained can also be used to calculate the desired trajectory of the vehicle. We evaluate LVD-NMPC against a baseline dynamic window approach (DWA) path planning executed using standard NMPC and against the PilotNet neural network. Performance is measured in our simulation environment GridSim, on a real-world 1:8 scaled model car as well as on a real size autonomous test vehicle and the nuScenes computer vision dataset.

A Vision-Based Lane Detection Approach for Autonomous Vehicles Using a Convolutional Neural Network Architecture

2021 ◽  
Author(s):  
Md. Al-Masrur Khan ◽  
Seong-Hoon Kee ◽  
Niloy Sikder ◽  
Md. Abdullah Al Mamun ◽  
Fatima Tuz Zohora ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Autonomous Vehicles ◽  
Network Architecture ◽  
Lane Detection ◽  
Neural Network Architecture ◽  
Detection Approach

Prediction of Vehicular Traffic Flow using Levenberg-Marquardt Artificial Neural Network Model: Italy Road Transportation System

2022 ◽  
Vol 24 (2) ◽  
pp. E74-E86
Author(s):  
Isaac Oyeyemi Olayode ◽  
Alessandro Severino ◽  
Tiziana Campisi ◽  
Lagouge Kwanda Tartibu
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Traffic Flow ◽  
Traffic Control ◽  
Traffic Congestion ◽  
Autonomous Vehicles ◽  
Time Of Day ◽  
Vehicle Speed ◽  
Road Transportation ◽  
Artificial Neural

In the last decades, the Italian road transport system has been characterized by severe and consistent traffic congestion and in particular Rome is one of the Italian cities most affected by this problem. In this study, a LevenbergMarquardt (LM) artificial neural network heuristic model was used to predict the traffic flow of non-autonomous vehicles. Traffic datasets were collected using both inductive loop detectors and video cameras as acquisition systems and selecting some parameters including vehicle speed, time of day, traffic volume and number of vehicles. The model showed a training, test and regression value (R2) of 0.99892, 0.99615 and 0.99714 respectively. The results of this research add to the growing body of literature on traffic flow modelling and help urban planners and traffic managers in terms of the traffic control and the provision of convenient travel routes for pedestrians and motorists.

Training Back-Propagation Neural Network for Target Localization Using Improved Particle Swarm Optimization

2013 ◽  
Vol 333-335 ◽  
pp. 1384-1387
Author(s):  
Jin Jie Yao ◽  
Xiang Ju ◽  
Li Ming Wang ◽  
Jin Xiao Pan ◽  
Yan Han
Keyword(s):  
Neural Network ◽  
Particle Swarm Optimization ◽  
Back Propagation ◽  
Particle Swarm ◽  
Back Propagation Neural Network ◽  
Position Estimation ◽  
Target Localization ◽  
Prediction Ability ◽  
Swarm Optimization ◽  
Improved Particle Swarm Optimization

Target localization technology has been intensively studied and broadly applied in many fields. This paper presents one improved particle swarm optimization technique in training a back-propagation neural network for position estimation in target localization. The proposed scheme combines particle swarm optimization (PSO), back-propagation neural network (BP), adaptive inertia weight and hybrid mutation, called IPSO-BP. To verify the proposed IPSO-BP approach, comparisons between the PSO-based BP approach (PSO-BP) and general back-propagation neural network (BP) are made. The computational results show that the proposed IPSO-BP approach exhibits much better performance in the training process and better prediction ability in the validation process than those using the other two base line approaches.

Sign in / Sign up

Export Citation Format

Share Document