Global Positioning System Reference
In-Depth Information
positioning system designed for moving environments and they carry out a key role in
hybrid positioning.
In this paper direct geo-referencing is achieved automatically from a mobile platform with
hybrid positioning by multi-sensor integration. Here, direct geo-referencing means geo-
referencing that does not require that the ground control points accurately measure ground
coordinate values. Data are acquired and digital surfaces are modeled using equipment
which is mounted on a mobile platform. This allows objects to be automatically rendered in
rich shapes and detailed textures.
2. System design for hybrid positioning and sensor integration
The key attributes of the design of the system are low cost, ease of use, and mobility (Parra
& Angel, 2005). Firstly, it utilizes a small laser scanner, commercially available digital
cameras, and a relatively inexpensive IMU such as FOG (Fiber Optic Gyro), not a high-
performance and expensive IMU like Ring Laser Gyro. The IMU and other measurement
tools used are much cheaper than those in existing aerial measurement systems, such as
Applanix's POS and Leica's ADS40 (Cramera, 2006). Moreover, these low-cost instruments
are easily available on the market. Recent technological advances have also led to low-cost
sensors such as micro electro mechanical system (MEMS) gyros. For example, it is
considered that MEMS gyros will supplant FOG in the near future and that the price will be
approximately one-tenth of that of FOG. For this reason, FOG was selected for this paper in
an attempt to improve a low-cost system for the future. Secondly, “mobility” here means the
item is lightweight and simple to modify. Such sensors allow the system to be borne by a
variety of platforms: UAV (Unmanned Aerial Vehicle), ground vehicles, humans, and
others. These sensors are generally low-performance, but they are light and low-cost while
still meeting the specifications. These handy sensors are improved by integrating their data.
2.1 Sensors
In this paper a laser scanner, digital cameras, an IMU, a GPS, and a VMS are used to find the
precise trajectory of sensors and to construct a digital surface model as a mobile mapping
system. To automatically construct such a model, it is necessary to develop a high-frequency
positioning system to determine the movement of the sensors in details. The integration of
GPS and IMU data is effective for high-accuracy positioning of a mobile platform. A 3D
shape is acquired by the laser scanner as point cloud data and texture information is
acquired by the digital cameras all from the same platform simultaneously. The sensors
used in this paper are listed in Table 1.
2.2 Sensors' calibration
Calibration of sensors is necessary for two reasons. One is to estimate the interior orientation
parameter, such as lens distortion and focal length that are mechanical oriented parameters.
The other reason calibration is necessary is to estimate exterior orientation parameters, such
as a transformation matrix that has a relative position and attitude among sensors. All the
sensors are tightly mounted on a platform and they have constant calibration parameters
during the measurement. The purpose of calibration is chiefly to integrate all the sensors
and positioning devices to a common single coordinating system, so that captured data can
be integrated and expressed in terms of a common world coordinate system.
Search WWH ::
Custom Search