US-20260125064-A1 - PREDICTION OF ROAD GRADE FOR AUTONOMOUS VEHICLE NAVIGATION

US20260125064A1US 20260125064 A1US20260125064 A1US 20260125064A1US-20260125064-A1

Abstract

Systems and methods of predicting a grade of a road upon which a vehicle is traveling are disclosed. An autonomous vehicle system can receive sensor data from a sensor measuring a response from at least one mechanical component of the autonomous vehicle as the autonomous vehicle navigates a road; detect a speed of the autonomous vehicle; determine a predicted grade of the road based on the sensor data and the speed; and navigate the autonomous vehicle based on the predicted grade of the road.

Inventors

Harish PULLAGURLA
Zachary Miller
Andrew Cunningham

Assignees

TORC ROBOTICS, INC.

Dates

Publication Date: 20260507
Application Date: 20251223

Claims (20)

1 . One or more non-transitory computer-readable storage media for predicting a road grade for an autonomous vehicle, the one or more non-transitory computer-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a system to: receive sensor data from at least one sensor measuring a response from at least one mechanical component of the autonomous vehicle while the autonomous vehicle navigates a road, the response including a power output of an engine of the autonomous vehicle; determine a speed of the autonomous vehicle; determine a first predicted grade of the road based on the power output and the speed; determine a second predicted grade of the road based on light detection and ranging (LiDAR) data captured by at least one LiDAR sensor of the autonomous vehicle; compare the first predicted grade with the second predicted grade; generate a confidence value of grade prediction based on the comparison; and navigate the autonomous vehicle based on the first predicted grade and the generated confidence value.
2 . The one or more non-transitory computer-readable storage media of claim 1 , wherein the plurality of instructions further cause the system to determine the first predicted grade by an artificial intelligence model, the artificial intelligence model trained with historical speed data and historical power output data.
3 . The one or more non-transitory computer-readable storage media of claim 1 , wherein the plurality of instructions further cause the system to: compare the first predicted grade with a third grade of the road, the third grade determined based on map data of the road; and transmit an indication that the third grade is incorrect in response to determining that a difference between the first predicted grade and the third grade being at or above a threshold.
4 . The one or more non-transitory computer-readable storage media of claim 1 , wherein the at least one sensor includes at least one of a rotational speed sensor, a torque sensor, a throttle position sensor, or a mass airflow sensor.
5 . The one or more non-transitory computer-readable storage media of claim 1 , wherein the plurality of instructions further cause the system to determine the first predicted grade, further based on a weight of the autonomous vehicle.
6 . The one or more non-transitory computer-readable storage media of claim 5 , wherein the plurality of instructions further cause the system to receive the weight of the autonomous vehicle from an external device.
7 . The one or more non-transitory computer-readable storage media of claim 1 , wherein the plurality of instructions further cause the system to transmit the first predicted grade to at least one remote server.
8 . The one or more non-transitory computer-readable storage media of claim 1 , the plurality of instructions further cause the system to: detect an object on the road by executing an object detection model, based on the first predicted grade.
9 . The one or more non-transitory computer-readable storage media of claim 1 , wherein the plurality of instructions further cause the system to: update map data of the autonomous vehicle, based on the first predicted grade.
10 . A computer-implemented method for predicting a road grade for an autonomous vehicle, the method comprising: receiving sensor data from at least one sensor measuring a response from at least one mechanical component of the autonomous vehicle while the autonomous vehicle navigates a road, the response including a power output of an engine of the autonomous vehicle; determining a speed of the autonomous vehicle; determining, by an artificial intelligence model, a predicted grade of the road based on the power output and the speed by: inputting the power output and the speed into the artificial intelligence model, the artificial intelligence model configured to output the predicted grade based on inputs to the artificial intelligence model, the predicted grade indicating steepness of the road; and navigating the autonomous vehicle based on the predicted grade.
11 . The method of claim 10 , wherein the artificial intelligence model is trained based on historical speed data and historical power output data captured by at least one vehicle traveling on roads of different grades.
12 . The method of claim 10 , wherein determining the predicted grade further comprises: determining the predicted grade by inputting a weight of the autonomous vehicle into the artificial intelligence model.
13 . The method of claim 10 , wherein the artificial intelligence model is trained by a remote server.
14 . An autonomy system of an autonomous vehicle, comprising at least one processor in communication with at least one memory, the at least one processor programmed to: receive sensor data from at least one sensor measuring a response from at least one mechanical component of the autonomous vehicle while the autonomous vehicle navigates a road, the response including a power output of an engine of the autonomous vehicle; determine a speed of the autonomous vehicle; determine, by an artificial intelligence model, a first predicted grade of the road based on the power output and the speed by: inputting the power output and the speed into the artificial intelligence model, the artificial intelligence model configured to output the first predicted grade based on inputs to the artificial intelligence model, the first predicted grade indicating steepness of the road; and navigate the autonomous vehicle based on the first predicted grade.
15 . The autonomy system of claim 14 , wherein the artificial intelligence model is trained based on historical speed data and historical power output data captured by at least one vehicle traveling on roads of different grades.
16 . The autonomy system of claim 14 , wherein the at least one processor is further programmed to: determine the first predicted grade by inputting a weight of the autonomous vehicle into the artificial intelligence model.
17 . The autonomy system of claim 14 , wherein the artificial intelligence model is trained by a remote server.
18 . The autonomy system of claim 14 , wherein the at least one processor is further programmed to: compare the first predicted grade with a third grade of the road, the third grade determined based on map data of the road; and transmit an indication that the third grade is incorrect in response to determining that a difference between the first predicted grade and the third grade being at or above a threshold.
19 . The autonomy system of claim 14 , wherein the at least one sensor includes at least one of a rotational speed sensor, a torque sensor, a throttle position sensor, or a mass airflow sensor.
20 . The autonomy system of claim 14 , wherein the at least one processor is further programmed to: determine a second predicted grade of the road based on light detection and ranging (LiDAR) data captured by at least one LiDAR sensor of the autonomous vehicle; compare the first predicted grade with the second predicted grade; generate a confidence value of grade prediction based on the comparison; and navigate the autonomous vehicle based on the first predicted grade and the generated confidence value.

Description

CROSS REFERENCE TO RELATED APPLICATIONS This application is a continuation application of U.S. patent application Ser. No. 18/222,384, filed on Jul. 14, 2023, the entire content and disclosure of which is hereby incorporated by reference herein in its entirety. TECHNICAL FIELD The present disclosure relates to autonomous vehicles and, more specifically, to predicting a grade of roads upon which autonomous vehicles are traveling. BACKGROUND The use of autonomous vehicles has become increasingly prevalent in recent years, with the potential for numerous benefits. One challenge faced by autonomous vehicles is modeling the surroundings of the autonomous vehicle. The grade of a road can affect fuel efficiency and energy expenditure, path planning, general navigation, and object detection capabilities of autonomous vehicles. Although conventional approaches based on satellite navigation systems, such as the global positioning system (GPS), may be utilized to estimate the grade of a road, such approaches return inaccurate results, particularly in urban environments where signal reception can be impaired. Similar issues with respect to accuracy occur when utilizing acceleration sensors such as inertial measurement units, because integration of small errors or noise in the IMU sensor data over time accumulate to produce inaccurate and unreliable estimations. These errors in road grade information may impair the ability of an autonomous vehicle to navigate properly. SUMMARY The systems and methods of the present disclosure may solve the problems set forth above and/or other problems in the art. The scope of the current disclosure, however, is defined by the attached claims, and not by the ability to solve any specific problem. Disclosed herein are techniques to automatically estimate the grade of a road for mapping and navigation purposes using the performance of an engine of a vehicle. The amount of power required by the engine to navigate the vehicle, while compensating for factors including vehicle load and wind resistance, can be utilized to determine an accurate approximation of an uphill or downhill gradient for the road upon which the vehicle is traveling. This is because the load caused by an uphill or downhill gradient will affect the amount of power required by the engine to automatically maintain the speed of the vehicle as it travels. Additionally, light detection and ranging (LiDAR) sensors as well as image sensors (e.g., cameras) may be utilized to estimate digital surface models of roads. For example, points generated by the LiDAR sensors and images captured by the image sensors can be utilized to generate the digital surface model as the vehicle is traveling on the road. The digital surface model may be utilized to assist with autonomous vehicle navigation, the generation of a world model or mapping data, and may be utilized in combination with the grade estimation techniques to more accurately predict the grade of the road upon which the vehicle is traveling. One embodiment is directed to a system. The system includes one or more processors of an autonomous vehicle. The system can receive sensor data from a sensor measuring a response from at least one mechanical component of the autonomous vehicle as the autonomous vehicle navigates a road; determine a speed of the autonomous vehicle; determine a predicted grade of the road based on the sensor data and the speed; and navigate the autonomous vehicle based on the predicted grade of the road. The sensor may comprise a rotational speed sensor, a torque sensor, or a throttle position sensor. The mechanical component may comprise an engine of the autonomous vehicle, and the one or more processors are further configured to determine a power output of the engine. The system may determine the predicted grade of the road based on the power output of the engine and the speed. The autonomous vehicle may comprise a LiDAR sensor. The system may determine the predicted grade of the road further based on LiDAR points captured by the LiDAR sensor. The system may determine the predicted grade of the road further based on a weight of the autonomous vehicle. The system may receive the weight of the autonomous vehicle from an external computing device. The system may transmit the predicted grade of the road to one or more remote servers. The system may execute an object detection model to detect an object on the road based on the predicted grade of the road. The system may update map data stored in memory of the autonomous vehicle based on the grade of the road. Another embodiment of the present disclosure is directed to a method. The method may be performed, for example, by one or more processors of an autonomous vehicle. The method includes receiving sensor data from a sensor measuring a response from at least one mechanical component of the autonomous vehicle as the autonomous vehicle navigates a road; determining a speed of the autonomous vehicle; determining a pred