Analysis of image-based navigation system for rotorcraft low-altitude flight

A synthetic vision system for enhancing the pilots ability to navigate and control the aircraft on the ground is described. The system uses the onboard airport database and images acquired by external sensors such as, millimeter wave, infra-red and low light TV cameras. Additional navigation information needed by the system is provided by the Inertial Navigation System and the Global Positioning System. The various functions of the system such as, image enhancement, map generation, obstacle detection, collision avoidance, guidance etc., are identified. The available technologies, some of which were developed at NASA that are applicable to the aircraft ground navigation problem, are noted. Example images of a truck crossing the runway while the aircraft flies close to the runway centerline are described. These images are from a sequence of images, acquired during one of the several flight experiments conducted by NASA to acquire data to be used for the development and verification of the synthetic vision concepts. These experiments provide a realistic database including video and infra-red images, motion states from the Inertial Navigation System and the Global Positioning System, and camera parameters.

Section: Obstacle Position and Velocity Estimationmentioning

confidence: 99%

Section: Obstacle Mapmentioning

confidence: 99%

<title>Computer-aided system for detecting runway incursions</title>

Sridhar

Chatterji

1994

“…The VPR however can exploit further parallelism at the ATU level if it contains more than one grid cell. 3 Each A TU/grid cell can be processed in parallel. The data requirements for each A TU are implicitly supplied by the parent V P R .…”

Section: Virtual Processing Regionsmentioning

confidence: 99%

“…The search window is constructed with knowledge of the imaging parameters, sensor geometry, a n d v ehicle state 7]. The peak of the correlation surface indicates the location of a feature at the new time, and can be used by an extended Kalman lter to estimate the spatial coordinates of the ground object which corresponds to the feature 3,6].…”

Section: Feature Tracking Algorithmmentioning

confidence: 99%

<title>Real-time computational needs of a multisensor feature-based range estimation method</title>

Suorsa

Sridhar

Fong³

1993

The computer vision literature describes many methods to perform obstacle detection and avoidance for autonomous or semi-autonomous vehicles. Methods may be broadly categorized into eld-based techniques and feature-based techniques. Field-based techniques have the advantage of regular computational structure at every pixel throughout the image plane. Feature-based techniques are much more data driven in that computational complexity increases dramatically in regions of the image populated by features. It is widely believed that to run computer vision algorithms in real time a parallel architecture is necessary. Field-based techniques lend themselves to easy parallelization due to their regular computational needs. However, we h a ve found that eld-based methods are sensitive to noise and have traditionally been di cult to generalize to arbitrary vehicle motion. Therefore, we h a ve sought t e c hniques to parallelize feature-based methods. This paper describes the computational needs of a parallel feature-based range-estimation method developed by NASA Ames. Issues of processing-element performance, load balancing, and data-ow bandwidth are addressed along with a performance review of two architectures on which the feature-based method has been implemented.

“…Several techniques have been proposed for range determination using electro-optical sensors, optical ow and state data from an inertial navigation system 6,7,8]. One algorithm of interest can detect, track and estimate range to image features (i.e., regions with common statistics or spatial structure) over time from a multisensor system mounted on a vehicle moving with arbitrary six degrees-of-freedom 9].…”

Section: Real-time Passive Range-estimationmentioning

confidence: 99%

<title>Real-time optical-flow range estimation on the iWarp</title>

Fong¹,

Suorsa

1993

This paper describes the implementation of an optical ow range-estimation algorithm, intended to support autonomous helicopter navigation, on the iWarp distributed-memory multicomputer. The implementation exploits numerous features of the iWarp communications and computation architecture to maximize performance and e ciency. The range-estimation algorithm has been previously implemented on both distributed and shared-memory workstations. Experiences and performance gures from the initial iWarp implementation are discussed along with plans for future re nements.