# US 2021/0080959 A1 · Target intention predicting method and system thereof

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0080959 A1
- Document ID: US20210080959A1
- Status: Published application
- Kind: A1
- Application number: 16/695,134
- Filed: 2019-11-25
- Published: 2021-03-18
- Earliest claimed: 2019-09-18
- Assignees: Automotive Research & Testing Center
- Applicants: Automotive Research & Testing Center
- Inventors: Wei-Jen WANG
- Web page: https://patentgrep.com/patent/US20210080959A1

## CPC

- G05D 1/0212: {with means for defining a desired trajectory (involving a plurality of land vehicles G05D1/0287)}
- G05D 1/0214: {in accordance with safety or protection criteria, e.g. avoiding hazardous areas (monitoring the location of vehicles within a certain area, e.g. forbidden or allowed areas, in traffic control systems for road vehicles G08G1/13)}
- G05D 1/0088: {characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00)}
- G05D 1/0274: {using mapping information stored in a memory device (navigation using map-matching G01C21/30)}
- G08G 1/166: {for active traffic, e.g. moving vehicles, pedestrians, bikes}
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G08G: TRAFFIC CONTROL SYSTEMS (guiding railway traffic, ensuring the safety of railway traffic B61L; arrangement of road signs or traffic signals E01F9/00; radar or analogous systems, sonar systems, lidar systems specially adapted for traffic control G01S13/91, G01S15/88, G01S17/88; {radar or analogous systems, sonar systems, lidar systems specially adapted for anti-collision purposes G01S13/93, G01S15/93, G01S17/93})

## Abstract

A target intention predicting method including a dataset obtaining step and a calculating and map mapping step is provided. A host vehicle positioning dataset of a host vehicle and a plurality of target datasets of a target are obtained. Each of the target datasets corresponds to each of a plurality of time points of a time line, and each of the target datasets includes a target position and a target velocity. The host vehicle positioning dataset is mapped to a map. The target position at the last one of the time points is mapped to the map. An updated velocity of the target is calculated according to the target velocities of the target datasets, and the updated velocity is mapped to the map to predicting a target future position of the target on the map at a future time point according to the updated velocity.

## Claims

### Claim 1

A target intention predicting method, comprising:
a dataset obtaining step to obtain a host vehicle positioning dataset of a host vehicle and a plurality of target datasets of at least one target, wherein each of the target datasets corresponds to each of a plurality of time points of a time line, and each of the target datasets comprises a target position and a target velocity; and
a calculating and map mapping step, comprising:
mapping the host vehicle positioning dataset to a map;
mapping the target position at the last one of the time points to the map; and
calculating an updated velocity of the at least one target according to the target velocities of the target datasets, and mapping the updated velocity to the map to predict a target future position of the at least one target on the map at a future time point according to the updated velocity.

### Claim 2 (depends on 1)

The target intention predicting method of claim 1, further comprising:
a future track choosing step, wherein the host vehicle comprises a plurality of future tracks on the map, and one of the future tracks is chosen according to the target future position of the at least one target such that the host vehicle evades the at least one target.

### Claim 3 (depends on 2)

The target intention predicting method of claim 2, wherein, in the calculating and map mapping step, a region of interest is adjusted according to the future tracks, and the updated velocity of the at least one target is calculated as long as the at least one target is within the region of interest.

### Claim 4 (depends on 3)

The target intention predicting method of claim 3, wherein a host vehicle future position of the future track corresponding to the last one of the future time points is found, a first setting distance is set between a front edge of the region of interest and the host vehicle future position, a second stetting distance is set between a rear edge of the region of interest and the host vehicle future position, a distance equal to a lane width is set between a left edge of the region of interest and the host vehicle future position, and another distance equal to the lane width is set between a left edge of the region of interest and the host vehicle future position.

### Claim 5 (depends on 1)

The target intention predicting method of claim 1, wherein the updated velocity is calculated according to the target velocities within a time threshold, the time threshold is defined to be the target velocity at the last one of the time points multiplied by an estimating value, and the estimating value is within 0 to 1.

### Claim 6 (depends on 5)

The target intention predicting method of claim 5, wherein the at least one target is classified into a vehicle type, a pedestrian-like type or a pedestrian type, and the estimating value of the vehicle type, the estimating value of the pedestrian-like type and the estimating value of the pedestrian type are different.

### Claim 7 (depends on 6)

The target intention predicting method of claim 6, wherein the at least one target is classified according to a position of the at least one target on the map, a length of the at least one target, a width of the at least one target, or a ratio of the length to the width.

### Claim 8 (depends on 7)

The target intention predicting method of claim 7, wherein the future time point is set to a value equal to the time threshold.

### Claim 9 (depends on 1)

The target intention predicting method of claim 1, wherein, in the calculating and map mapping step, the updated velocity is calculated by a weighted moving average method.

### Claim 10 (depends on 1)

The target intention predicting method of claim 1, wherein, in the calculating and map mapping step, the host vehicle positioning dataset is obtained by a real time kinematic method.

### Claim 11 (depends on 1)

A target intention predicting system, which is applied to the target intention predicting method of claim 1, comprising:
the host vehicle;
at least one sensor disposed at the host vehicle to detect the at least one target; and
a processor signally connected to the at least one sensor to obtain the target datasets of the at least one target.

### Claim 12 (depends on 11)

The target intention predicting system of claim 11, wherein the least one sensor has an optical sensor structure.
