Center for Mapping




System Architecture and Description

 

 

Positioning Module

 

The key hardware components of the current implementation of AIMSTM include two dual-frequency Trimble 4000SSI GPS receivers and a medium-accuracy and high reliability strapdown Litton LN-100 inertial navigation system, based on the Zero-lockTM Laser Gyro (ZLGTM) and an A-4 accelerometer triad (0.8 nmi/h CEP, gyro bias – 0.003°/h, accelerometer bias – 25mg). The LN-100 firmware, modified for the AIMSTM project, allows for access to the raw IMU data, updated at 256 Hz. In the AIMS solution, a single Kalman filter estimates errors in position, velocity, and attitude, as well as errors in inertial and GPS measurements. It implements GPS L1/L2 phase observables in double-differenced mode together with the INS strapdown navigation solution. The tight integration architecture implemented in AIMS allows the feedback of estimated IMU errors to the INS navigation module to improve INS accuracy, and support a high rate (256 Hz) for navigation output.

 

A flexible architecture for AIMS (Figure 1) enables augmentation of a variety of sensors beyond high-resolution CCD (Charge-Coupled Device) cameras, including Synthetic Aperture Radar (SAR), radar or laser ranging devices, to support the Image Acquisition Module. The GPS time and the 1 PPS signal are used to synchronize position information with measurements from the other sensors. The ultimate goal is to build a real- or near real-time system. Currently, however, AIMS operates in a post-processing mode. Unlike conventional aerial photography, the system requires limited ground control, namely one or two base stations, to perform differential GPS. By eliminating the need for ground control, analytical aerotriangulation, and analog to digital conversion, AIMS produces spatial data at a much lower cost than conventional aerial mapping methods in use today.

 

 

 


Figure 1. AIMS prototype hardware configuration, 1997.

 


GPS/INS integration

GPS and INS, as navigation techniques, offer highly complementary operational characteristics by using entirely different positioning principles – a radio navigation satellite system, GPS provides essentially geometric information while autonomous INS offers inertial information, i.e., the reaction to the applied force. GPS in a stand-alone mode provides a position fix as long as it is able to maintain lock to a minimum of four satellites. Some systems are still able to provide a solution with less than four satellites, but the accuracy is substantially degraded, especially if partial loss of lock is quite extensive in time. Naturally, one way to improve the GPS gap bridging is to add an autonomous, passive navigation sensor, such as INS. Most of the modern MMS systems rely on high-accuracy differential GPS and quality strapdown INS, while early systems used simpler, and lower quality dead-reckoning systems (wheel counter or odometer, directional and vertical gyros). Naturally, the accuracy obtained strongly depends on the type of sensors used, and ranges from meters (early systems) to centimeters. Inertial navigation systems provide self-contained and independent means of three?dimensional positioning and orientation with potentially high short-term accuracy. In addition, compared to conventional GPS output rate, INS provides much higher positioning update rates (up to 256 Hz). However, INS accuracy degrades over time due to uncompensated gyro and accelerometer errors. Thus, with full operational GPS capability, it has been recognized that an optimal combination of GPS with inertial navigation brings a number of advantages over stand-alone inertial or GPS navigation. GPS contributes its high accuracy and stability over time, enabling continuous monitoring of inertial sensor errors. Implementation of closed-loop INS error calibration in Kalman filter environment allows continuous, on-the-fly error update (and thus INS calibration), leading to increased estimation accuracy. The effective error level depends on systematic and random effects on the GPS measurements, as amplified by satellite geometry. Well-calibrated, GPS-supported INS provides precise position and attitude information between the GPS updates and during GPS losses of lock, facilitating immunity to GPS outages, continuous attitude solution, and reduction of the GPS ambiguity search volume/time, after the signal reacquisition. In general, using a GPS-calibrated, high to medium accuracy inertial system, attitude accuracy in the range of 10-30 arcsec can be achieved. In summary, any combination of GPS and INS functionality into a single integrated navigation system represents a fusion of dissimilar, complementary data, and should be able to provide a superior performance as opposed to either sensor in a stand-alone mode. In fact, integration of these two systems is often the only way to achieve the following goals.



Imaging Module

 

The AIMS imaging component consists of a digital camera, based on a 4,096 by 4,096 CCD with a 60 mm by 60 mm imaging area (15-micron pixel size), manufactured by Lockheed Martin Fairchild Semiconductors.

 

The 4K´4K imaging sensor is integrated into a camera-back (BigShot) of a regular Hasselblad 553 ELX camera body together with a supporting data acquisition interface. The complete digital camera system installed on a rigid mount together with the INS is shown in Figure 2. The Hasselblad 553 ELX (Figure 3) features an electronic control system providing the necessary functionality for a fully digital, computer-controlled camera operation. Zeiss CF lenses with 50 and 80 mm focal lengths supplement the experimental camera system, offering wide and normal angle configurations. It is a high-resolution digital camera, which can be controlled remotely by a host computer through a SCSI-2 interface.

High-resolution digital imaging sensors have been primarily studied in academia and in the military/reconnaissance community. Although currently available CCD resolution is still modest in comparison to the 20K by 20K-equivalent resolution of quality aerial cameras, there are already numerous applications where existing digital imaging sensors can easily satisfy project requirements. Furthermore, electronic imaging is a precondition for any real-time application. Moreover, digital imaging sensors have a clear advantage over analog film-based techniques by substantially reducing turnaround time and providing easy connections to softcopy systems.

 

 


 


               Figure 2. AIMS hardware configuration                           Figure 3. BigShotTM Hasselblad camera

                           – INS and digital  camera mount.

 

 

System Architecture and Description| GPS/INS Integration and Kalman Filtering| Performance Evaluation| Ambiguity Resolution