Dissertation > Excellent graduate degree dissertation topics show

Research on Multi-Objective Path Planning and Track Control of Unconstrained Mobile

Author: WangGuangBin
Tutor: TianYanTao
School: Jilin University
Course: Control Theory and Control Engineering
Keywords: Environmental modeling Multi-objective planning Cricket system Evaluation function Fuzzy Optimization
CLC: TP242
Type: Master's thesis
Year: 2011
Downloads: 54
Quote: 0
Read: Download Dissertation

Abstract

Cricket system is a typical nonlinear control systems, domestic and foreign scholars in the cricket system there is a certain, in a flat above the ball due to its own movement is to rely on the movement of the plate to produce, so we moving body is referred to as unconstrained. Cricket system is a system of very representative of a class of nonlinear control and motion planning of multi-objective. Study Cricket multi-objective motion planning process, we will be in a tablet is seen as the top of the camera is an external perception system, real-time images collected through the camera to get the current environmental information, to conduct multi-objective motion planning problem, and the traditional path planning problems such as robot path planning and vehicle navigation systems have a lot of similarities, but it is due to the characteristics of its mechanical structure, the ball itself is not subject to mechanical constraints , therefore they have differences, this is cricket system research another problem, namely the problem of trajectory tracking control. In this paper, the typical nonlinear system, cricket system target motion planning and trajectory tracking control problem. Study the problem of multi-objective motion planning process, many scholars at home and abroad have adopted intelligent algorithms such as fuzzy and neural network in trajectory tracking control like a fuzzy algorithm has been implemented. Cricket system for laboratory study multi-objective motion planning process involves the following work: 1. Environment model and the location of the ball extracted. Multi-objective motion planning problem for cricket system in this article is an off-line planning, and its essence is to first plan out the right path, and then track. Therefore, we must first solve the problem of environment recognition. Upcoming obstacles, traps and the position of the ball in the environment map identified before planning. This paper shows the image histogram characteristics on the basis of existing research laboratory experience many experiments, the trap pixel first determine 30 first trap position to determine, and then use the histogram binarization processing, to extract the location of the ball, while determining the position of the obstacle in three peaks of the two peak and valley pixels. Environment map while the extraction position rasterized processed at the same time the trap grid position is marked as -1, the the obstacle grid position marked 1, the the free grid position marked as 0, motion planning for the next multi-objective for preparation. Unconstrained movement of the body more than the target motion planning establishment and expression. In multi target motion planning, multi-objective problem is one of the more important issues. , Laboratory research work in the earlier stage of expression for the path length in the multi-objective, dangerous and time is not given a mathematical model of the degree of risk, and time is assumed to be uniform for planning. In this paper, this shortfall, given the different obstacles and traps hazards models, and the introduction of a metric function to measure the size of the time, so that the path length, risk and time respectively precise mathematical modeling. Unified physical quantities on three paper introduced multi-objective fuzzy optimization method to determine the final evaluation function, simultaneously solving multi-objective motion planning process, the path to the final will be in accordance with the minimum evaluation function direction on the search until the target point. Trajectory tracking control algorithm. According to the characteristics of multi-target motion planning of cricket system, combined with the PID parameters on the characteristics of the control system. Using the fuzzy PID controller to calculate the flat desired rotation angle and the actual angle of rotation differential input to calculate the number of pulses of the motor needs to rotate in the inner ring of a PD controller, and then drive motor rotation, so that the small balls along track with planning out the path until the end. Cricket system experimental platform to achieve multi-objective motion planning. Environmental map information obtained after the acquisition to image processing, artificial given the weights of the multi-objective problem, in accordance with the need for the allocation of weights for multi-objective motion planning path search, more in line with actual requirements and people subjective desire. The experimental procedure found in different operating conditions, the system planning of the path is different. Experiment, respectively, for the environment, the right to the same value and the same environment, different weights, final results show that this multi-goal motion planning method is better to find a different path to meet the different requirements of the people, some are shorter path length , some higher safety factor, while others may be the shortest time. In this case depends on the weights of the people selected. Different weights will produce different path. The experimental study also established cricket the system competition evaluation index system, a preliminary study of the minimum energy path. Establish cricket the system competition evaluation index system for different experiments, gives a standard of comparison exactly which experiments better. The cricket the system competition evaluation index system established based on the completion of the experiment is good or bad, such as the size of the path length, the length of time consumption and tracking process whether a trap or tripping over obstacles. Minimum energy path is also the problem of path planning one of the hot spots for the minimum energy path system physical structure of cricket and force a simple study of the energy consumption of the formula, according to the ultimate source of energy using the integral method to strike and simple experimental study. In summary, the main work of this paper is to study the obstacles and traps in the complex environment with obstacles and traps the cricket system of multi-target motion planning problem for this complex environment, different risk modeling, while introducing a measure expressed as a function of the size of the time, In addition to the use of multi-objective fuzzy optimization method and multi-objective motion planning cricket system evaluation function is given explicit mathematical expression and solving process of trajectory tracking controller simultaneously established a system of cricket contest evaluation index system, a preliminary study of the minimum energy path, carried out a series of comparative experimental study in different operating conditions.

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Robotics > Robot
© 2012 www.DissertationTopic.Net Mobile