US 12,286,103 B2
Apparatus, method, and computer program for object detection
Daisuke Toyoda, Tokyo-to (JP); and Junya Inada, Kariya (JP)
Assigned to TOYOTA JIDOSHA KABUSHIKI KAISHA, Toyota (JP)
Filed by TOYOTA JIDOSHA KABUSHIKI KAISHA, Toyota (JP)
Filed on Aug. 17, 2021, as Appl. No. 17/404,036.
Claims priority of application No. 2020-141086 (JP), filed on Aug. 24, 2020.
Prior Publication US 2022/0055618 A1, Feb. 24, 2022
Int. Cl. G06F 18/24 (2023.01); B60W 30/095 (2012.01); B60W 40/04 (2006.01); B60W 40/105 (2012.01); G06V 20/56 (2022.01)
CPC B60W 30/095 (2013.01) [B60W 40/04 (2013.01); B60W 40/105 (2013.01); G06F 18/24 (2023.01); G06V 20/56 (2022.01); B60W 2554/4046 (2020.02); B60W 2554/80 (2020.02)] 8 Claims
OG exemplary drawing
 
1. An apparatus for object detection, comprising:
a processor configured to:
detect an object around a vehicle by inputting a sensor signal into a classifier, the sensor signal being obtained by at least one sensor mounted on the vehicle for sensing the situation around the vehicle;
set the degree of omission of computation so that at least part of computation on the sensor signal by the classifier will be omitted when the speed of the vehicle is not greater than a predetermined speed threshold and the situation around the vehicle does not correspond to a first situation in which the number of objects around the vehicle is not less than a predetermined number, a second situation in which the distance between the vehicle and an object around the vehicle is not greater than a predetermined distance, a third situation in which the speed of an object around the vehicle relative to the vehicle is not less than a predetermined threshold, a fourth situation in which the vehicle is stuck in a traffic jam, nor a fifth situation in which the vehicle is stopped beyond a stop line;
generate a planned trajectory based on an output of the classifier; and
control the vehicle to travel along the planned trajectory; wherein
the processor omits part of computation on the sensor signal by the classifier according to the set degree of omission of computation so that a size of a feature map which is calculated by a predetermined layer of the classifier when the computation on the sensor signal by the classifier is omitted is smaller than the size of the feature map when the computation on the sensor signal by the classifier is not omitted; and
when the speed of the vehicle is not greater than the predetermined speed threshold and the situation around the vehicle corresponds to the fourth or the fifth situation but does not correspond to any of the first to the third situations, the processor sets the degree of omission of computation so that the cycle in which the sensor signal is inputted into the classifier will be longer than the cycle for the case that the speed of the vehicle is greater than the predetermined speed threshold.