Gear based vehicle load inference system
AssigneeBaidu USA LLC
InventorFan Zhu
Life of this patent
Estimating the term…
Abstract
According to various embodiments, described herein are methods and systems for collecting data for training a load inference regression model for use in an ADV. According to one exemplary method, an ADV is manually driven on a segment of a road for a number of periods of time. During each period of time, a set of features of the ADV are recorded, including one or more features at a first time prior to a gear shift from a first gear position to a second gear position, and one or more features at a second time after the gear shift. For each of the number of periods of time, a weight of the ADV is also recorded using a weight sensor. The recorded features and the total weight of the ADV for each of the periods of time are then used to train a neural network regression model for inferring a load of the ADV in real time.
Claims20
A computer-implemented method of collecting data for training a load inference model for use in an autonomous driving vehicle (ADV), the method comprising:
- recording, by the ADV, a set of driving statistics of the ADV during each of a plurality of trips on a road segment, wherein the ADV is manually driven on the road segment and has a different load during each of the plurality of trips, wherein the ADV performs a plurality of gear shifts during each of the plurality of trips;
- recording, by the ADV, a total weight of the ADV during each of the plurality of trips;
- extracting, from the set of driving statistics, a plurality of features of the ADV associated with each of the plurality of gear shifts during each of the plurality of trips, including one or more features at a first time prior to each gear shift from a first gear position a second gear position and one or more features at a second time after the gear shift, wherein the plurality of features are utilized to train the load inference model, wherein the load inference model is a neural network model and is to determine a load of the ADV to dynamically adjust actuator commands of the ADV.
A non-transitory machine-readable medium having instructions stored therein for collecting data for training a load inference model for use in an autonomous driving vehicle (ADV), which instructions, when executed by a processor, cause the processor to perform operations, the operations comprising:
- recording, by the ADV, a set of driving statistics of the ADV during each of a plurality of trips on a road segment, wherein the ADV is manually driven on the road segment and has a different load during each of the plurality of trips, wherein the ADV performs a plurality of gear shifts during each of the plurality of trips;
- recording, by the ADV, a total weight of the ADV during each of the plurality of trips;
- extracting, from the set of driving statistics, a plurality of features of the ADV associated with each of the plurality of gear shifts during each of the plurality of trips, including one or more features at a first time prior to each gear shift from a first gear position to a second gear position and one or more features at a second time after the gear shift, wherein the plurality of features are utilized to train the load inference model, wherein the load inference model is a neural network model and is to determine a load of the ADV to dynamically adjust actuator commands of the ADV.
A data processing system, comprising:
- a processor;
- a memory coupled to the processor and storing program instructions for training a load inference model for use in an autonomous driving vehicle (ADV), which, when executed by the processor, cause the processor to perform operations comprising:
- recording, by the ADV, a set of driving statistics of the ADV during each of a plurality of trips on a road segment, wherein the ADV is manually driven on the road segment and has a different load during each of the plurality of trips, wherein the ADV performs a plurality of gear shifts during each of the plurality of trips;
- recording, by the ADV, a total weight of the ADV during each of the plurality of trips;
- extracting, from the set of driving statistics, a plurality of features of the ADV associated with each of the plurality of gear shifts during each of the plurality of trips, including one or more features at a first time prior to each gear shift from a first gear position to a second gear position and one or more features at a second time after the gear shift, wherein the plurality of features are utilized to train the load inference model, wherein the load inference model is a neural network model and is to determine a load of the ADV to dynamically adjust actuator commands of the ADV.
Dependencies are inferred from the claim text.