Neural Network Based Low Cost Autonomous Vehicle
نویسندگان
چکیده
The conceptual high-level design of a low cost neural network (NN) based autonomous vehicle is presented. Focus is on the domain of indoor as well as small outdoor applications such as mail delivery system among neighboring organizations. The design is based on four independent neural networks attached with ultrasonic rangefinders; Inertial Measuring Unit (IMU) and GPS receiver. The NN model is the multilayer feed-forward network with back-propagation training algorithm. The NN implementation is done using the Atmel RISC microcontrollers (MC). The absolute referencing is achieved with the combination of GPS receiver and IMU. A vector map is incorporated for road following and path planning. The arbitration scheme is based on fuzzy logic and implemented with the same series of MC. The low cost components make the design feasible for our daily use applications.
منابع مشابه
Calibration of an Inertial Accelerometer using Trained Neural Network by Levenberg-Marquardt Algorithm for Vehicle Navigation
The designing of advanced driver assistance systems and autonomous vehicles needs measurement of dynamical variations of vehicle, such as acceleration, velocity and yaw rate. Designed adaptive controllers to control lateral and longitudinal vehicle dynamics are based on the measured variables. Inertial MEMS-based sensors have some benefits including low price and low consumption that make them ...
متن کاملScheduling of Multiple Autonomous Guided Vehicles for an Assembly Line Using Minimum Cost Network Flow
This paper proposed a parallel automated assembly line system to produce multiple products having multiple autonomous guided vehicles (AGVs). Several assembly lines are configured to produce multiple products in which the technologies of machines are shared among the assembly lines when required. The transportation between the stations in an assembly line (intra assembly line) and among station...
متن کاملModeling and Intelligent Control System Design for Overtaking Maneuver in Autonomous Vehicles
The purpose of this study is to design an intelligent control system to guide the overtaking maneuver with a higher performance than the existing systems. Unlike the existing models which consider constant values for some of the effective variables of this behavior, in this paper, a neural network model is designed based on the real overtaking data using instantaneous values for variables. A fu...
متن کاملDesign and Experimental Evaluation of integrated orientation estimation algorithm Autonomous Underwater Vehicle Based on Indirect Complementary Filter
This paper aims is to design an integrated navigation system constituted by low-cost inertial sensors to estimate the orientation of an Autonomous Underwater Vehicle (AUV) during all phases of under water and surface missions. The proposed approach relied on global positioning system, inertial measurement unit (accelerometer & rate gyro), magnetometer and complementary filter technique. Complem...
متن کاملAggressive Deep Driving: Model Predictive Control with a CNN Cost Model
We present a framework for vision-based model predictive control (MPC) for the task of aggressive, high-speed autonomous driving. Our approach uses deep convolutional neural networks to predict cost functions from input video which are directly suitable for online trajectory optimization with MPC. We demonstrate the method in a high speed autonomous driving scenario, where we use a single monoc...
متن کامل