Handling uncertainty associated with a sensor measurement using a previously determined combined occupancy grid

1 . A system for handling uncertainty associated with a sensor measurement using a previously determined combined occupancy grid, the system comprising:

a plurality of sensors; and

an electronic processor, the electronic processor configured to,

determine a combined occupancy grid associated with a first time using measurements received from the plurality of sensors and associated with the first time;

receive a measurement associated with a second time from a sensor of the plurality of sensors;

determine an uncertainty grid associated with the second time including an uncertainty distribution associated with the measurement associated with the second time;

convolute the combined occupancy grid associated with the first time with the uncertainty grid associated with the second time to determine one or more convolution values, wherein each convolution value is associated with a cell in the uncertainty grid;

determine a cell associated with a greatest convolution value;

assign a first probability to the cell associated with the greatest convolution value;

assign a second probability to a cell associated with a mean of the uncertainty distribution;

using the first probability and the second probability, determine a combined occupancy grid associated with the second time; and

control a machine based on the combined occupancy grid associated with the second time, wherein the machine is configured to move autonomously or semi-autonomously.

2 . The system according to claim 1 , wherein the uncertainty distribution is a non-linear distribution or a uniform distribution.

3 . The system according to claim 2 , wherein the non-linear distribution is a Gaussian ellipse.

4 . The system according to claim 1 , wherein the measurement is of a location of an external object within an environment of the machine and the mean of the uncertainty distribution is a most likely location of the external object based on the measurement from the sensor.

5 . The system according to claim 1 , wherein the first probability is the greatest convolution value and the second probability is 1 minus the greatest convolution value.

6 . The system according to claim 1 , wherein the machine is a vehicle and the electronic processor is configured to control the vehicle using one or more vehicle control systems.

7 . The system according to claim 1 , wherein each cell of the uncertainty grid that is included in the uncertainty distribution is assigned a distribution value and each cell that is included in the combined occupancy grid associated with the first time is assigned an occupancy value.

8 . The system according to claim 1 , wherein the electronic processor is configured to convolute the combined occupancy grid associated with the first time with the uncertainty grid to determine one or more convolution values, wherein each convolution value is associated with a cell in the uncertainty grid by

for each cell of a plurality of cells included in the uncertainty grid and the uncertainty distribution,

determining a distribution value assigned to the cell;

determining an occupancy value assigned to a cell included in the combined occupancy grid associated with the first time that is associated with the cell included in the uncertainty grid; and

assigning the cell included in the uncertainty grid a convolution value, wherein the convolution value is equal to the distribution value multiplied by the occupancy value.

9 . The system according to claim 1 , wherein the electronic processor is configured to convolute the combined occupancy grid associated with the first time with the uncertainty grid to determine one or more convolution values, wherein each convolution value is associated with a cell in the uncertainty grid by

for each cell included in the uncertainty grid and within a predetermined range from the mean of the uncertainty distribution,

determining a distribution value assigned to the cell;

determining an occupancy value assigned to a cell included in the combined occupancy grid associated with the first time that is associated with the cell included in the uncertainty grid; and

assigning the cell included in the uncertainty grid a convolution value, wherein the convolution value is equal to the distribution value multiplied by the occupancy value.

10 . The system according to claim 1 , wherein the electronic processor is configured to, using the first probability and the second probability, determine a combined occupancy grid associated with the second time by

multiplying the first probability and the second probability by a weight;

for each cell included in the combined occupancy grid associated with the second time,

when the cell of the combined occupancy grid associated with the second time is associated with a cell included in the combined occupancy grid associated with the first time, assign a mass value of the cell included in combined occupancy grid associated with the first time to the cell of the combined occupancy grid associated with the second time; and

when the cell of the combined occupancy grid associated with the second time is a new cell, assign a mass value of 0 to the cell of the combined occupancy grid associated with the second time; and

updating the mass values assigned to cells included in the combined occupancy grid associated with the second time using the weighted first probability and the weighted second probability.

11 . The system according to claim 10 , wherein the weight is a value representing a likelihood an external object associated with the measurement exists.

12 . The system of claim 1 , wherein the electronic processor is further configured to

for each cell in the uncertainty grid associated with a convolution value,

determine whether the convolution value is above a predetermined threshold; and

when the convolution value is above the predetermined threshold, endorse the cell; and

using the first probability and the second probability and the endorsements, determine a combined occupancy grid associated with the second time.

13 . The system of claim 1 , wherein the sensor is a radar sensor, a LiDAR sensor, or a camera.

14 . A method for handling uncertainty associated with a sensor measurement using a previously determined combined occupancy grid, the method comprising:

determining a combined occupancy grid associated with a first time using measurements received from a plurality of sensors and associated with the first time;

receiving a measurement associated with a second time from a sensor of the plurality of sensors;

determining an uncertainty grid associated with the second time including an uncertainty distribution associated with the measurement associated with the second time;

convoluting the combined occupancy grid associated with the first time with the uncertainty grid associated with the second time to determine one or more convolution values, wherein each convolution value is associated with a cell in the uncertainty grid;

determining a cell associated with a greatest convolution value;

assigning a first probability to the cell associated with the greatest convolution value;

assigning a second probability to a cell associated with a mean of the uncertainty distribution;

using the first probability and the second probability, determining a combined occupancy grid associated with the second time; and

controlling a machine based on the combined occupancy grid associated with the second time, wherein the machine is configured to move autonomously or semi-autonomously.

15 . The method according to claim 14 , wherein the uncertainty distribution is a non-linear distribution or a uniform distribution.

16 . The method according to claim 15 , wherein the non-linear distribution is a Gaussian ellipse.

17 . The method according to claim 14 , wherein the machine is a vehicle and controlling a machine based on the combined occupancy grid associated with the second time includes controlling the vehicle using one or more vehicle control systems.

18 . The method according to claim 14 , wherein convoluting the combined occupancy grid associated with the first time with the uncertainty grid to determine one or more convolution values, wherein each convolution value is associated with a cell in the uncertainty grid includes

for each cell of a plurality of cells included in the uncertainty grid and the uncertainty distribution,

determining a distribution value assigned to the cell;

determining an occupancy value assigned to a cell included in the combined occupancy grid associated with the first time that is associated with the cell included in the uncertainty grid; and

assigning the cell included in the uncertainty grid a convolution value, wherein the convolution value is equal to the distribution value multiplied by the occupancy value.

19 . The method according to claim 14 , wherein convoluting the combined occupancy grid associated with the first time with the uncertainty grid to determine one or more convolution values, wherein each convolution value is associated with a cell in the uncertainty grid includes

for each cell included in the uncertainty grid and within a predetermined range from the mean of the uncertainty distribution,

determining a distribution value assigned to the cell;

determining an occupancy value assigned to a cell included in the combined occupancy grid associated with the first time that is associated with the cell included in the uncertainty grid; and

assigning the cell included in the uncertainty grid a convolution value, wherein the convolution value is equal to the distribution value multiplied by the occupancy value.

20 . The method of claim 14 , the method further comprising

for each cell in the uncertainty grid associated with a convolution value,

determining whether the convolution value is above a predetermined threshold; and

when the convolution value is above the predetermined threshold, endorsing the cell; and

using the first probability and the second probability and the endorsements, determining a combined occupancy grid associated with the second time.