Event classification and object tracking
First Claim
1. A method comprising:
- assigning, by a computing system, a pre-classification to a detection event in an environmental model having an environmental coordinate field that spans multiple non-overlapping measurement coordinate fields of sensors mounted in a vehicle, wherein the detection event corresponds to spatially-aligned and temporally-aligned raw sensor measurements indicative of an object proximate to the vehicle, and wherein the environmental model is stored by the computing system and includes both the spatially-aligned and temporally-aligned raw sensor measurements and an indication of the detection event;
updating, by the computing system, the environmental model to include new raw sensor measurements from the sensors mounted in the vehicle and corresponding detection events over time; and
tracking, by the computing system, the detection event in the updated environmental model based on the assigned pre-classification of the detection event, wherein a control system for the vehicle is configured to control operation of the vehicle based, at least in part, on the tracked detection event.
3 Assignments
0 Petitions
Accused Products
Abstract
This application discloses a computing system to implement object tracking in an assisted or automated driving system of a vehicle. The computing system can assign a pre-classification to a detection event in an environmental model, update the environmental model with new sensor measurements and corresponding detection events over time, and track the detection event in the updated environmental model. The computing system can track the detection event by predicting a future state of the detection event with a state change model selected based on the assigned pre-classification, comparing the predicted future state to an actual future state of the detection event in an update to the environmental model, and determining the detection event corresponds to an object proximate to the vehicle based on the comparison. A control system for the vehicle can control operation of the vehicle based, at least in part, on the tracked detection event.
-
Citations
20 Claims
-
1. A method comprising:
-
assigning, by a computing system, a pre-classification to a detection event in an environmental model having an environmental coordinate field that spans multiple non-overlapping measurement coordinate fields of sensors mounted in a vehicle, wherein the detection event corresponds to spatially-aligned and temporally-aligned raw sensor measurements indicative of an object proximate to the vehicle, and wherein the environmental model is stored by the computing system and includes both the spatially-aligned and temporally-aligned raw sensor measurements and an indication of the detection event; updating, by the computing system, the environmental model to include new raw sensor measurements from the sensors mounted in the vehicle and corresponding detection events over time; and tracking, by the computing system, the detection event in the updated environmental model based on the assigned pre-classification of the detection event, wherein a control system for the vehicle is configured to control operation of the vehicle based, at least in part, on the tracked detection event. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. An apparatus comprising at least one memory device storing instructions configured to cause one or more processing devices to perform operations comprising:
-
assigning a pre-classification to a detection event in an environmental model having an environmental coordinate field that spans multiple non-overlapping measurement coordinate fields of sensors mounted in a vehicle, wherein the detection event corresponds to spatially-aligned and temporally-aligned raw sensor measurements indicative of an object proximate to the vehicle, and wherein the environmental model is stored by the memory device and includes both the spatially-aligned and temporally-aligned raw sensor measurements and an indication of the detection event; updating the environmental model to include new raw sensor measurements from the sensors mounted in the vehicle and corresponding detection events over time; and tracking the detection event in the updated environmental model based on the assigned pre-classification of the detection event, wherein a control system for the vehicle is configured to control operation of the vehicle based, at least in part, on the tracked detection event. - View Dependent Claims (9, 10, 11, 12, 13, 14)
-
-
15. A system comprising:
-
a memory device configured to store machine-readable instructions; and a computing system including one or more processing devices, in response to executing the machine-readable instructions, configured to; assign a pre-classification to a detection event in an environmental model having an environmental coordinate field that spans multiple non-overlapping measurement coordinate fields of sensors mounted in a vehicle, wherein the detection event corresponds to spatially-aligned and temporally-aligned raw sensor measurements indicative of an object proximate to the vehicle, and wherein the environmental model is stored by the memory device and includes both the spatially-aligned and temporally-aligned raw sensor measurements and an indication of the detection event; update the environmental model to include new raw sensor measurements from the sensors mounted in the vehicle and corresponding detection events over time; and track the detection event in the updated environmental model based on the assigned pre-classification of the detection event, wherein a control system for the vehicle is configured to control operation of the vehicle based, at least in part, on the tracked detection event. - View Dependent Claims (16, 17, 18, 19, 20)
-
Specification