A comparison between low-cost and professional RTK equipment under forest conditions
Typ
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Program
Wireless, photonics and space engineering (MPWPS), MSc
Publicerad
2024
Författare
Wan, Yiqing
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
Real time kinematic (RTK) is an important differential GNSS positioning method.
It is widely used in forestry investigation. Depending on the type of reference station,
RTK technique can be divided into traditional RTK and the advanced network RTK.
There are also different types of GNSS receivers, for example the low-cost mapping
type and the professional surveying type. The surveying type devices tend to have
more accurate results, but the price difference between the two types of receivers
can be dozens of times, and surveying type receivers are usually too bulky to carry
and move around in the forest. Therefore, it is worthwhile to analyze if there are
any situations that low-cost traditional RTK devices can achieve a comparable per formance as the professional network RTK devices do.
In this project, we compared the performance of a low-cost mapping type receiver,
u-blox F9P, with a professional network RTK device, Leica GS18 I. Field measure ments were taken in two days in the forests near Gothenburg. Measurements were
taken in different percentages of canopy closure, different canopy directions and
different tree species. Standard deviations and positional errors of the collected po sitions were calculated. Variables like dilution of precision (DOP), age of differential
(AoD), sky plot and time to fix were also collected.
It is found that the real time correction messages (RTCM) communication range for
u-blox was at least 500 m in flat area, but terrain obstacles might reduce the range
limit by half. The u-blox device performed bad when the canopy closure was over
50% or 65%, with a positional error of over 50 cm, thus it is not recommended to
replace Leica network RTK with u-blox in dense forests. In most cases the standard
deviation of Leica was better than that of u-blox. DOP was an important factor
that had strong correlations with errors, fix status and cycle slips. Data collection
with u-blox should wait for 1∼5 minutes after power-on, until DOP becomes lower
and stable. AoD is found not much related with the error, but there should not be
obvious break with the communication of RTCM. Canopy directions influenced the
error but had less impact than canopy closure. The relationship between the error
and tree species was not able to be determined.
Beskrivning
Ämne/nyckelord
GNSS, real time kinematic, network RTK, forest, canopy closure, dilution of precision, age of differential