Pranav Maheshwari

US Patent:

20200209858, Jul 2, 2020

Filed:

Mar 6, 2019

Appl. No.:

16/294274

Inventors:

- Orlando FL, US
Pranav Maheshwari - Palo Alto CA, US
Vahid R. Ramezani - Portola Valley CA, US

International Classification:

G05D 1/00
G06N 20/00
G06N 3/08
G05D 1/02

Abstract:

To generate a machine learning model for controlling autonomous vehicles, training sensor data is obtained from sensors associated with one or more vehicles, the sensor data indicative of physical conditions of an environment in which the one or more vehicles operate, and a machine learning (ML) model is trained using the training sensor data. The ML model generates parameters of the environment in response to input sensor data. A controller in an autonomous vehicle receives sensor data from one or more sensors operating in the autonomous vehicle, applies the received sensor data to the ML model to obtain parameters of an environment in which the autonomous vehicle operates, provides the generated parameters to a motion planner component to generate decisions for controlling the autonomous vehicle, and causes the autonomous vehicle to maneuver in accordance with the generated decisions.

US Patent:

20200043176, Feb 6, 2020

Filed:

Nov 20, 2018

Appl. No.:

16/196630

Inventors:

- Orlando FL, US
Pranav Maheshwari - Palo Alto CA, US
Benjamin Englard - Palo Alto CA, US

International Classification:

G06T 7/246
G06T 7/521
G06K 9/00
G06T 7/00
G06T 7/73
G01S 17/58
G06T 5/00

Abstract:

A computer-implemented method of determining relative velocity between a vehicle and an object. The method includes receiving sensor data generated by one or more sensors of the vehicle configured to sense an environment by following a scan pattern comprising component scan lines. The method includes obtaining, based on the sensor data, a point cloud frame. Additionally, the method includes identifying a first pixel and a second pixel that are co-located within a field of regard and overlap a point cloud object within the point cloud frame and calculating a difference between a depth associated with the first pixel and a depth associated with the second pixel. The method includes determining a relative velocity of the point cloud object by dividing the difference in depth data by a time difference between when the depth associated with the first pixel was sensed and the depth associated with the second pixel was sensed.

US Patent:

20190113920, Apr 18, 2019

Filed:

Oct 2, 2018

Appl. No.:

16/149225

Inventors:

- Orlando FL, US
Pranav Maheshwari - Palo Alto CA, US
Shubham C. Khilari - Palo Alto CA, US
Vahid R. Ramezani - Portola Valley CA, US

International Classification:

G05D 1/00
G06N 3/08
G06N 5/02
G05D 1/02
G01C 21/34

Abstract:

A computer-readable medium stores instructions executable by one or more processors to implement a self-driving control architecture for controlling an autonomous vehicle. A perception component receives sensor data and generates signals descriptive of a current state of the environment. Based on those signals, a prediction component generates signals descriptive of one or more predicted future environment states. A motion planner generates decisions for maneuvering the vehicle toward a destination, at least by using the signals descriptive of the current and future environment states to set values of one or more independent variables in an objective equation. The objective equation includes terms corresponding to different driving objectives over a finite time horizon. Values of one or more dependent variables in the objective equation are determined by solving the equation subject to a set of constraints, and values of the dependent variables are used to generate decisions for maneuvering the vehicle.

US Patent:

20190113927, Apr 18, 2019

Filed:

Oct 2, 2018

Appl. No.:

16/149223

Inventors:

- Orlando FL, US
Gauri Gandhi - Mountain View CA, US
Pranav Maheshwari - Palo Alto CA, US

International Classification:

G05D 1/02
G06N 3/08
G06N 5/02
G05D 1/00
G01C 21/34

Abstract:

A computer-readable medium stores instructions executable by one or more processors to implement a self-driving control architecture for controlling an autonomous vehicle. A perception and prediction component receives sensor data, and generates (1) an observed occupancy grid indicating which cells are currently occupied in a two-dimensional representation of the environment, and (2) predicted occupancy grids indicating which cells are expected to be occupied later. A mapping component provides navigation data for guiding the vehicle toward a destination, and a cost map generation component is configured to generate, based on the observed occupancy grid, the predicted occupancy grid(s), and the navigation data, cost maps that each specify numerical values representing a cost, at a respective instance of time, of occupying certain cells in a two-dimensional representation of the environment. A motion planner generates a grid path through the environment based on the cost maps, and corresponding decisions for maneuvering the vehicle.