Calculate Motion Statistics

Available in real-time and big data analytics.

The Calculate Motion Statistics tool calculates motion statistics and measures for observations based on the key fields specified for the Track ID and Start Time options. This tool is found in the Data Enrichment tool folder. Calculations include distance, timespan, height, speed, acceleration, heading, slope, and idling based on time values from the Start Time option and geometry of point-based events.

Example

A city is monitoring snowplow operations and city officials want to better understand vehicle movement. The Calculate Motion Statistics tool can be used to determine idle locations and time spent idling, average and maximum speeds, average distance traveled between observations, and other statistics.

Usage notes

Keep the following in mind when working with the Calculate Motion Statistics tool:

• The Calculate Motion Statistics tool has a state store depth value to specify the value for the Motion statistics are calculated using the last <number of observations> observations per track parameter. This value determines the number of observations the tool caches per track in the state store. As new observations for a track are received, older observations of the same track are deleted. The tool uses the observations in the state store to calculate motion statistics. For example, if the state store depth is set to five observations per track, the current observation and previous four observations are used to calculate statistics such as minimum acceleration, maximum speed, and average distance traveled per observation.
  • Three observations per track are required to calculate acceleration values, so this parameter must be set at a positive integer value greater than or equal to three.
  • Not all motion statistics are calculated using all observations in the state store. For example, the idling statistic is calculated using only the current observation and the previous observation.
• The tool outputs all input observations. As a result, the first observation for each track contains null values for motion statistics because there are no previous observations in the state store to use for calculations. Until there are enough observations in the track to meet the state store depth value, motion statistics are calculated using the available observations from the state store. Once there are enough observations to meet the state store depth value, motion statistics are calculated using the number of observations specified.
• In real-time analytics, the Calculate Motion Statistics tool uses the maximum time to retain observations value to set the value for the Maximum amount of time to retain observations for a track parameter. A previous observation is deleted from the state store if the sum of its Start Time value and the value specified for this parameter is less than the Start Time value of the current observation.
  • For example, if a value of one hour is used, all previous observations with a Start Time value that is not within an hour of the current observation's Start Time values are deleted from the state store. If the state store depth is set to 4, and the current observation has a Start Time value of 12/10/2025 08:00:00 and the previous observations in the state store have Start Time values of 12/10/2025 06:59:00, 12/10/2025 07:00:00, and 12/10/2025 07:30:00, the observation with a Start Time value of 12/10/2025 06:59:00 is deleted from the state store. The other observations are retained because their Start Time values are within one hour of the current observation’s start time.
  • If this parameter deletes observations from the state store and there are less than two observations per track remaining, no statistics can be calculated for the current observation. At least three observations per track are required to calculate motion statistics related to acceleration.
• This tool uses two values for distance and time to determine whether an observation is idling: idle distance tolerance and idle timespan tolerance. In ArcGIS Velocity, these appear as Observations are idling if they have moved less than or equal to <idle distance tolerance (distance units)> and more than <idle time tolerance (time units)> has elapsed since the last observation.
  • Idling is set to True if the distance between the current observation and the previous observation is less than or equal to the idle distance tolerance value and the duration between the current observation and the previous observation is greater than the idle time tolerance value. Idling is set to False if one or both conditions are not met.
  • For example, if the idle distance tolerance value is set to 50 meters and the idle time tolerance value is set to 5 seconds, an observation would be considered as idling if the distance between the current and previous observation was 45 meters and the duration between the Start Time values of the current and previous observations was 6 seconds. If the duration between the current and previous observations was 5 seconds instead, the feature would not be considered idling, since only the idle distance tolerance condition is met.
• The unit of measurement for the output motion statistics depends on the unit of measurement specified for the idle distance tolerance and idle timespan tolerance values. For example, if you specify an idle distance tolerance value of 750 meters and the idle timespan tolerance value of 10 minutes, the output distance fields are in meters, the output timespan fields are in minutes, and the output speed fields contain values with a meters-per-minute unit of measurement.
• The Method parameter defines how distances and other spatial calculations are made.
  • Planar—Euclidean calculations are generated. Euclidean distance is measured in a two-dimensional Cartesian plane in which a straight line is calculated between two points on a flat surface, otherwise known as the Cartesian plane. Euclidean distance is the more common type of distance calculation and works well when analyzing distances around features in a projected coordinate system that are concentrated in a relatively small area such as one UTM zone. Euclidean distances are referred to as planar distances.
  • Geodesic—Geodesic calculations are generated. Geodesic distances account for the actual shape of the earth, an ellipsoid, or more properly, a geoid. Distances are calculated between two points on a curved surface, the geoid, as opposed to two points on a flat surface, the Cartesian plane. Consider using geodesic calculations in the following circumstances:
    • The input features are dispersed covering multiple UTM zones, large regions, or the entire globe.
    • The spatial reference, or map projection, of the input features distorts distances to preserve other properties such as area.
• In real-time analytics, the tool operates in a stateful manner, allowing it to compare sequential observations to one another to calculate motion statistics—for example, how the current position, speed, altitude, and other statistics are different from that of the previous observation.
• In real-time analytics, the Calculate Motion Statistics tool maintains a state store of the current observation for each track ID. The state store depth parameter determines the size of the state store per track. When a new observation is received, it is compared to the current observation of the same track ID, if present, to calculate motion attributes. If the new observation has a newer timestamp than the current observation in the state store, the current observation is replaced. If it has an identical timestamp to the current observation for the same track ID, the tool does not distinguish between the two observations. This means the tool does not replace the current observation in the state store with the new observation, the new observation is not added to the state store, and future observations for this track ID are compared to the current observation that remains in the state store until a new observation arrives with a later timestamp. A warning is logged by the tool when this occurs.
• In big data analytics, one or more fields can be specified to identify tracks in addition to the one specified for the Track ID option. Tracks are represented by the unique combination of one or more track fields.

Parameters

The following are required parameters for the Calculate Motion Statistics tool:

Parameter	Description	Data type
Input Layer	The point observations for which motion statistics are calculated. This dataset must have key fields defined for the Track ID and Start Time options.	Features
Motion statistics are calculated using the last <number of observations> observations per track (state store depth)	The number of cached observations per track in the state store, including the current observation. Motion statistics are calculated based on the observations in the state store. For example, if you set a state store depth value of 3, each time an observation is processed, motion statistics calculations such as average speed are calculated based on the speed of the current observation and the previous two observations. Not all motion statistics are calculated using all observations in the state store, such as idling, which uses only the current observation and the one before it. Note: The first observation for each track has null values for motion statistics, as there are no previous observations in the state store to use for calculations. Until there are enough observations in the track to meet the state store depth value, motion statistics are calculated using all available observations from the state store. Once there are enough observations to meet the state store depth value, motion statistics are calculated using the number of observations specified. For example, if you set the state store depth value to 3, the first observation for each track has null motion statistics (based on the key field specific in the Track ID option). The second observation has motion statistics calculated based on the current and previous observations. For all following observations, motion statistics are calculated from available observations (three observations per track).	Integer
Maximum amount of time to retain observations for a track (maximum time to retain observations) (real-time analytics only)	A previous observation will be deleted from the state store if the sum of its Start Time value and the value specified for this parameter is less than the Start Time value of the current observation. For example, if a value of 1 hour is used, all observations with a start time that is not within an hour of the current observation's start time are deleted from the state store. Caution: If this parameter deletes observations from the state store and there are less than two observations per track remaining, no statistics can be calculated for the current observation. At least three observations per track are required for calculating acceleration statistics.	String
Observations are idle if they have moved more than <time units> have elapsed since the last observation (idle timespan tolerance)	The idle timespan tolerance value is used to determine idling. Idling is set to True if the distance between the current observation and the previous observation is less than or equal to the idle distance tolerance value and the duration between the current observation and the previous observation is greater than the idle time tolerance value. Idling is set to False if one or both of these conditions are not met.	String
Observations are idle if they have moved less than or equal to <distance units> have elapsed since the last observation (idle distance tolerance)	The idle distance tolerance value is used to determine idling. Idling is set to True if the distance between the current observation and the previous observation is less than or equal to the idle distance tolerance value and the duration between the current observation and the previous observation is greater than the idle time tolerance value. Idling is set to False if one or both of these conditions are not met.	String
Method	Specifies whether the distance calculation method is Geodesic or Planar (Euclidean). Note: If Planar is chosen, the input points must be projected. You can use the Project tool to project the input features before calculating distances. You do not need to use the tool if the input points are already projected.	String Allowed values: Planar | Geodesic
Track ID Fields (big data analytics only)	The fields that uniquely identify features to be processed. If a track ID field is specified, the Track ID value is automatically set on the Input Layer value.	Fields

Output layer

The output layer contains the attributes described below appended to the original attributes for each feature. Some statistics are calculated using only the previous and current observations while others use all observations in the state store.

Note:

The unit of measurement for the output motion statistics depends on the unit of measurement specified for the idle distance tolerance and idle timespan tolerance values. For example, if you specify an idle distance tolerance value of 750 meters and an idle timespan tolerance value of 10 minutes, the output distance fields are in meters, the output timespan fields are in minutes, and the output speed fields contain values with a meters-per-minute unit of measurement.

Field name	Description	Calculated using	Field type
Distance	The distance traveled since the previous observation.	Current and previous observations	Float64
TotalDistance	The total distance traveled from the oldest observation in the state store to the current observation.	All state store observations	Float64
MinDistance	The minimum distance traveled between observations.	All state store observations	Float64
MaxDistance	The maximum distance traveled between observations.	All state store observations	Float64
AvgDistance	The average distance traveled between observations.	All state store observations	Float64
Speed	The current speed since the previous observation.	Current and previous observations	Float64
TotalSpeed	The cumulative speed.	All state store observations	Float64
MinSpeed	The minimum speed.	All state store observations	Float64
MaxSpeed	The maximum speed.	All state store observations	Float64
AvgSpeed	The average speed.	All state store observations	Float64
TimeSpan	The time elapsed since the previous observation.	Current and previous observations	Float64
TotalTimeSpan	The total time elapsed from the oldest observation in the state store to the current observation.	All state store observations	Float64
MinTimeSpan	The minimum time elapsed between observations.	All state store observations	Float64
MaxTimeSpan	The maximum time elapsed between observations.	All state store observations	Float64
AvgTimeSpan	The average timespan between observations.	All state store observations	Float64
Height	The current height of the observation.	Current observation	Float64
HeightChange	The change in height from the previous observation.	Current and previous observations	Float64
TotalHeightChange	The total change in height from the oldest observation in the state store to the current observation. Note: This can be a negative value.	All state store observations	Float64
MinHeight	The minimum height observed.	All state store observations	Float64
MaxHeight	The maximum height observed.	All state store observations	Float64
AvgHeight	The average (mean) height observed.	All state store observations	Float64
Slope	The current slope of travel since the previous observation.	Current and previous observations
MinSlope	The minimum slope.	All state store observations	Float64
MaxSlope	The maximum slope.	All state store observations	Float64
AvgSlope	The average slope. Note: Average slope is calculated based on the slopes observed since the oldest observation in the state store. If the state store depth value is 3, the average slope is the average of the slope from observation 1 to observation 2 and from observation 2 to observation 3.	All state store observations	Float64
Acceleration	The acceleration since the previous observation.	Current and previous observations, requires that speed has been calculated for the previous observation	Float64
TotalAcceleration	The total acceleration from the oldest observation in the state store to the current observation.	All state store observations	Float64
MinAcceleration	The minimum acceleration.	All state store observations	Float64
MaxAcceleration	The maximum acceleration.	All state store observations	Float64
AvgAcceleration	The average acceleration. Note: Average acceleration is calculated based on the accelerations observed since the oldest observation in the state store. If the state store depth value is 3, average acceleration is the average acceleration from observation 1 to observation 2, and from observation 2 to observation 3.	All state store observations	Float64
Idling	Specifies whether the track is idle based on the distance between the current and previous observations and timespan between the current and previous observations.	Current and previous observations	Boolean
TotalIdleTime	The cumulative amount of time during which idling was detected since the oldest observation in the state store.	All state store observations	Float64
PercentageIdleTime	The percentage of time during which idling was detected since the oldest stored observation in the state store.	All state store observations	Float64
Heading	The geodetic bearing. This is an angle of the travel direction (geodetic bearing, which is the angle measured clockwise from north).	Current and previous observations	Float64