Object tracking device and object tracking method
An object tracking device and object tracking method that can track an object accurately without increasing the computational load are provided. An object tracking device ( 20 ) includes an input interface ( 21 ), a processor ( 23 ), and an output interface ( 24 ). The input interface ( 21 ) is configured to acquire sensor data. The processor ( 23 ) is configured to detect a detection target from the sensor data and track the detection target using Kalman filters associated with each of the detection target and an observed value. The output interface ( 24 ) is configured to output a detection result regarding the detection target. The processor ( 23 ) is configured to execute a first process to select a Kalman filter with low certainty from among a plurality of the Kalman filters associated with the same detection target or observed value as an exclusion candidate Kalman filter that could be excluded from the association, and a second process to exclude from the association and initialize the exclusion candidate Kalman filter that meets an initialization condition.
1 . An object tracking device comprising:
an input interface configured to acquire sensor data,
a processor configured to detect a detection target from the sensor data and track the detection target using Kalman filters associated with each of the detection target and an observed value, and
an output interface configured to output a detection result regarding the detection target,
wherein the processor is configured to execute
a first process to select a Kalman filter with low certainty from among a plurality of the Kalman filters in an association with a same detection target or an observed value as an exclusion candidate Kalman filter that can be excluded from the association, and
a second process to exclude from the association and initialize the exclusion candidate Kalman filter that meets an initialization condition, and
wherein the processor is configured to execute the first process and the second process when a number of Kalman filters associated with the same detection target or the same observed value exceeds a second value.
2 . The object tracking device according to claim 1 , wherein the processor is configured to determine a certainty of a Kalman filter according to a size of an error ellipse.
3 . The object tracking device according to claim 1 , wherein the initialization condition is that the Kalman filter with low certainty is selected as the exclusion candidate Kalman filter a number of times equal to a first value.
4 . The object tracking device according to claim 1 , wherein the processor is configured to execute the first process and the second process in a process of associating the Kalman filters with the observed value.
5 . The object tracking device according to claim 1 , wherein the processor is configured to execute the first process and the second process in a process of associating the Kalman filters with the detection target.
6 . An object tracking method comprising:
acquiring sensor data,
detecting a detection target from the sensor data and tracking the detection target using Kalman filters associated with each of the detection target and an observed value, and
outputting a detection result regarding the detection target,
wherein tracking of the detection target includes
a first process to select a Kalman filter with low certainty from among a plurality of the Kalman filters in an association with a same detection target or observed value as an exclusion candidate Kalman filter that can be excluded from the association, and
a second process to exclude from the association and initialize the exclusion candidate Kalman filter that meets an initialization condition, and
wherein tracking of the detection target includes executing the first process and the second process when a number of Kalman filters associated with the same detection target or the same observed value exceeds a second value.