Font Size: a A A

Elevation Terrain Reconstruction In Front Of Vehicles Research Based On Gaussian Distribution

Posted on:2021-01-19Degree:MasterType:Thesis
Country:ChinaCandidate:C J WangFull Text:PDF
GTID:2392330611471274Subject:Engineering
Abstract/Summary:
Emergency rescue vehicles usually drive in complex terrain environments.The preview active suspension technology has an important role in maintaining the vehicle speed.The reconstruction of the terrain environment in front of the vehicle is a prerequisite for achieving the preview active control.In order to ensure the accuracy and real-time of vehicle control,the real-time reconstruction problem applicable to the terrain environment in front of heavy vehicles must be solved.This subject is derived from the China’s National Key R&D Programmes "Research on Key Chassis and Suspension Technology for High-Mobility Emergency Rescue Vehicles(Including fire trucks)"(Project No.:2016YFC0802902),explores the positioning methods and 3D terrain reconstruction methods in front of the vehicle of unstructured roads for outdoor emergency rescue vehicles based on LiDAR(Light Detection and Ranging),INS(Inertial Navigation System),and GNSS(Global positioning and navigation system).This article mainly includes the following:According to the error characteristics of LiDAR,INS,and GNSS,Li DAR outlier elimination method,INS acceleration unwanted component elimination method and GNSS position accuracy improvement method are designed and experimental verification are completed.The complementary characteristics of GNSS and INS in positioning are researched.From the perspective of improving positioning accuracy and frequency,a combined positioning method model based on EKF(Extended Kalman Filter)is established.Simulation experiments show that EKF can effectively improve positioning accuracy.According to the characteristics of the INS sensor when the GNSS signal weakens,a height restricted EKF combined positioning method is designed.The characteristics of OctoMap(occupied map)and elevation map environment representation methods are researched.Through comparative simulation experiments,it is shown that elevation map has better mapping effects in terms of memory occupation,robustness and continuity.The effects of elevation map resolution on the expression of map information were analyzed through simulation experiments,and appropriate resolution parameters were selected.A mobile test vehicle was built,and the combined positioning method based on EKF and the 3D reconstruction method of elevation terrain were used to verify the positioning and mapping effects through slope tests,positioning tests with weakened GNSS signals,and speed belt tests during straight and turning.
Keywords/Search Tags:emergency rescue vehicle, extended kalman filter, integrated positioning, terrain environment in front of the vehicle, real-time reconstruction, elevation map
Related items