Implementation and Metrics for a Trajectory Prediction Validation Methodology

Paglione, Mike; Oaks, Robert D.

doi:10.2514/6.2007-6517

Cited by 21 publications

(7 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The remaining 78,220 trajectories were used for testing. Based on the recommendations provided in [32,55,56], we measure the following three error metrics for the aircraft trajectory prediction experiment. The trajectory prediction errors are calculated for each observed radar track point in each input trajectory segment.…”

Section: Experiments 1: Terminal Area Air Traffic Predictionmentioning

confidence: 99%

Tree Memory Networks for modelling long-term temporal dependencies

et al. 2018

View full text Add to dashboard Cite

In the domain of sequence modelling, Recurrent Neural Networks (RNN) have been capable of achieving impressive results in a variety of application areas including visual question answering, part-of-speech tagging and machine translation. However this success in modelling short term dependencies has not successfully transitioned to application areas such as trajectory prediction, which require capturing both short term and long term relationships. In this paper, we propose a Tree Memory Network (TMN) for jointly modelling both long term relationships between multiple sequences and short term relationships within a sequence, in sequence-to-sequence mapping problems. The proposed network architecture is composed of an input module, controller and a memory module. In contrast to related literature which models the memory as a sequence of historical states, we model the memory as a recursive tree structure. This structure more effectively captures temporal dependencies across both short and long term time periods through its hierarchical structure. We demonstrate the effectiveness and flexibility of the proposed TMN in two practical problems: aircraft trajectory modelling and pedestrian trajectory modelling in a surveillance setting. In both cases the proposed approach outperforms the current state-of-the-art. Furthermore, we perform an in depth analysis on the evolution of the memory module content over time and provide visual evidence on how the proposed TMN is able to map both short and long term relationships efficiently via a hierarchical structure.

show abstract

Section: Experiments 1: Terminal Area Air Traffic Predictionmentioning

confidence: 99%

Tree Memory Networks for modelling long-term temporal dependencies

et al. 2018

View full text Add to dashboard Cite

show abstract

“…We will break down each variable so the reader is fully aware of how we determine the strengths and weaknesses of a TP [12].…”

Section: Trajectory Predictor Comparisonmentioning

confidence: 99%

Enhanced Star Glyphs for Multiple-Source Data Analysis

Rusu

Santiago²,

Crowell³

et al. 2009

2009 13th International Conference Information Visualisation

View full text Add to dashboard Cite

The analysis of large sums of data can be extremely difficult to perform if the data is not presented graphically. As a result, many graphing techniques have been developed, such as scatter plots, histograms. Generally, the main purpose of graphically displaying data is to do one of two things: First, to find the general average of where most of the data lies. Second, to find the outliers, the data points that are most distant from the others. Our visualization will attempt to find both by using a multitude of common graphing techniques to expand upon the traditional star glyph and create a new way of graphing data. These techniques include clustering, using color as identifiers, and 3D graphing capabilities to present more data that would not be possible of being shown in a two dimensional environment. We apply our techniques to compare several air traffic trajectory predictors currently being analyzed by the U.S. Federal Aviation Administration.

show abstract

“…Figure 4 provides an illustration of the horizontal deviation, and details on the calculation of this metric can be found in "Implementation and Metrics for a Trajectory Prediction Validation Methodology." [10] I1-7…”

Section: Weather Reroute Analysismentioning

confidence: 99%

Modeling weather in simulation and analysis

Young¹,

Crowell²,

Fabian³

2013

2013 Integrated Communications, Navigation and Surveillance Conference (ICNS)

View full text Add to dashboard Cite

In the concept development and validation process, fast-time simulation and modeling exercises are performed to examine system performance, obtain initial assessments of potential benefits, and to identify potential problem areas where real-time simulation studies are necessary for further exploration [1]. Despite the immense impact weather has on the National Airspace System (NAS), most simulation and analysis tools are unable to represent weather activity. This creates a large gap in the capabilities of future concept analysis such that potential benefits of some operational improvements cannot be quantified. The FAA's Concept Analysis Branch (ANG-C41) developed a tool which uses high fidelity weather data to create weather polygons. These weather polygons provide a much more concise model to store and process in simulation and analysis tools. In a fast-time simulation environment, the weather polygons can be modeled as restricted airspace that moves across the NAS. This enables researchers to measure the impact of operational improvements on weather-related flight delays. The weather polygons can also be used in analytical tools such as ANG-C41's FliteViz4D. Incorporating the weather polygons into this tool allows the user to animate both air traffic and weather in one threedimensional view. This is very useful when examining the impact of weather on air traffic. In this paper, the Concept Analysis Branch describes the process of creating and using weather polygons for simulation and analysis activities. An example study is discussed where weather polygons are used to determine the impact of weather on flight efficiency in today's NAS. Finally, the weather polygons are tested in a fast-time simulation environment.

show abstract

Implementation and Metrics for a Trajectory Prediction Validation Methodology

Cited by 21 publications

References 2 publications

Tree Memory Networks for modelling long-term temporal dependencies

Tree Memory Networks for modelling long-term temporal dependencies

Enhanced Star Glyphs for Multiple-Source Data Analysis

Modeling weather in simulation and analysis

Contact Info

Product

Resources

About