# US 2019/0317512 A1 · Method to evaluate trajectory candidates for autonomous driving vehicles (advs)

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

- Number: US 2019/0317512 A1
- Document ID: US20190317512A1
- Status: Published application
- Kind: A1
- Application number: 15/955,584
- Filed: 2018-04-17
- Published: 2019-10-17
- Earliest claimed: 2018-04-17
- Assignees: Baidu USA LLC
- Applicants: Baidu USA LLC
- Inventors: Yajia Zhang; Kecheng Xu
- Web page: https://patentgrep.com/patent/US20190317512A1

## CPC

- B60W 60/00: Drive control systems specially adapted for autonomous road vehicles
- G05D 1/0217: {in accordance with energy consumption, time reduction or distance reduction criteria}
- B60W 30/095: Predicting travel path or likelihood of collision
- B60W 60/0011: {involving control alternatives for a single driving scenario, e.g. planning several paths to avoid obstacles}
- B60W 60/0013: {specially adapted for occupant comfort}
- G05D 1/0238: {using obstacle or wall sensors (G05D1/0246 and G05D1/0289 take precedence; lidar systems designed for anti-collision purposes G01S17/93)}
- B60W 2520/10: Longitudinal speed
- B60W 2520/105: Longitudinal acceleration
- B60W 2556/50: of positioning data, e.g. GPS [Global Positioning System] data
- B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES

## Abstract

In one embodiment, a system generates a plurality of trajectory candidates for an autonomous driving vehicle (ADV) from a starting point to an end point of a particular driving scenario. The system generates a reference trajectory corresponding to the driving scenario based on a current state of the ADV associated with the starting point and an end state of the ADV associated with the end point, where the reference trajectory is associated with an objective. For each of the trajectory candidates, the system compares the trajectory candidate with the reference trajectory to generate an objective cost representing a similarity between the trajectory candidate and the reference trajectory. The system selects one of the trajectory candidates as a target trajectory for driving the ADV based on objective costs of the trajectory candidates.

## Claims

### Claim 1

A computer-implemented method to determining a trajectory to operate an autonomous driving vehicle, the method comprising:
generating a plurality of trajectory candidates for an autonomous driving vehicle (ADV) from a starting point to an end point of a particular driving scenario;
generating a reference trajectory corresponding to the driving scenario based on a current state of the ADV associated with the starting point and an end state of the ADV associated with the end point, wherein the reference trajectory is associated with an objective;
for each of the trajectory candidates, comparing the trajectory candidate with the reference trajectory to generate an objective cost representing a similarity between the trajectory candidate and the reference trajectory; and
selecting one of the trajectory candidates as a target trajectory for driving the ADV based on objective costs of the trajectory candidates.

### Claim 2 (depends on 1)

The method of claim 1, wherein comparing the trajectory candidate with the reference trajectory comprises:
segmenting the trajectory candidate into a plurality of candidate segments;
segmenting the reference trajectory into a plurality of reference segments;
for each of the candidate segments, calculating a velocity difference between a velocity of the candidate segment and a velocity of a corresponding reference segment; and
calculating the objective cost based on the velocity differences between the candidate segments and reference segments.

### Claim 3 (depends on 2)

The method of claim 2, wherein the objective cost is calculated by summing squared velocity differences between the trajectory segments and the reference segments.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
for each of the trajectory candidates,
identifying an obstacle that is closest to the trajectory candidate,
measuring a distance between the trajectory candidate and the obstacle,
calculating a safety cost associated with the trajectory candidate based on the distance between the trajectory candidate and the obstacle, and
calculating a total trajectory cost for the trajectory candidate based on the objective cost and the safety cost; and
selecting the target trajectory from the trajectory candidates, wherein the target trajectory having a lowest total trajectory cost.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
for each of the trajectory candidates,
determining a changing rate of acceleration along the trajectory candidate,
calculating a comfort cost for the trajectory candidate based on the changing rate of acceleration along the trajectory candidate, and
calculating a total trajectory cost for the trajectory candidate based on the objective cost and the comfort cost; and
selecting the target trajectory from the trajectory candidates, wherein the target trajectory having a lowest total trajectory cost.

### Claim 6 (depends on 5)

The method of claim 5, wherein calculating a comfort cost for the trajectory candidate comprises:
segmenting the trajectory candidate into a plurality of candidate segments;
for each of the candidate segments, calculating a segment comfort cost for the candidate segment based on a changing rate of an acceleration associated with the candidate segment; and
calculating the comfort cost based on the segment comfort costs of the candidate segments.

### Claim 7 (depends on 6)

The method of claim 6, wherein the comfort cost is calculated based on a sum of segment comfort costs of the candidate segments.

### Claim 8

A non-transitory machine-readable medium having instructions stored therein, which when executed by one or more processors, cause the one or more processors to perform operations, the operations comprising:
generating a plurality of trajectory candidates for an autonomous driving vehicle (ADV) from a starting point to an end point of a particular driving scenario;
generating a reference trajectory corresponding to the driving scenario based on a current state of the ADV associated with the starting point and an end state of the ADV associated with the end point, wherein the reference trajectory is associated with an objective;
for each of the trajectory candidates, comparing the trajectory candidate with the reference trajectory to generate an objective cost representing a similarity between the trajectory candidate and the reference trajectory; and
selecting one of the trajectory candidates as a target trajectory for driving the ADV based on objective costs of the trajectory candidates.

### Claim 9 (depends on 8)

The non-transitory machine-readable medium of claim 8, wherein comparing the trajectory candidate with the reference trajectory comprises:
segmenting the trajectory candidate into a plurality of candidate segments;
segmenting the reference trajectory into a plurality of reference segments;
for each of the candidate segments, calculating a velocity difference between a velocity of the candidate segment and a velocity of a corresponding reference segment; and
calculating the objective cost based on the velocity differences between the candidate segments and reference segments.

### Claim 10 (depends on 9)

The non-transitory machine-readable medium of claim 9, wherein the objective cost is calculated by summing squared velocity differences between the trajectory segments and the reference segments.

### Claim 11 (depends on 8)

The non-transitory machine-readable medium of claim 8, further comprising:
for each of the trajectory candidates,
identifying an obstacle that is closest to the trajectory candidate,
measuring a distance between the trajectory candidate and the obstacle,
calculating a safety cost associated with the trajectory candidate based on the distance between the trajectory candidate and the obstacle, and
calculating a total trajectory cost for the trajectory candidate based on the objective cost and the safety cost; and
selecting the target trajectory from the trajectory candidates, wherein the target trajectory having a lowest total trajectory cost.

### Claim 12 (depends on 8)

The non-transitory machine-readable medium of claim 8, further comprising:
for each of the trajectory candidates,
determining a changing rate of acceleration along the trajectory candidate,
calculating a comfort cost for the trajectory candidate based on the changing rate of acceleration along the trajectory candidate, and
calculating a total trajectory cost for the trajectory candidate based on the objective cost and the comfort cost; and
selecting the target trajectory from the trajectory candidates, wherein the target trajectory having a lowest total trajectory cost.

### Claim 13 (depends on 12)

The non-transitory machine-readable medium of claim 12, wherein calculating a comfort cost for the trajectory candidate comprises:
segmenting the trajectory candidate into a plurality of candidate segments;
for each of the candidate segments, calculating a segment comfort cost for the candidate segment based on a changing rate of an acceleration associated with the candidate segment; and
calculating the comfort cost based on the segment comfort costs of the candidate segments.

### Claim 14 (depends on 13)

The non-transitory machine-readable medium of claim 13, wherein the comfort cost is calculated based on a sum of segment comfort costs of the candidate segments.

### Claim 15

A data processing system, comprising:
one or more processors; and
a memory coupled to the one or more processors to store instructions, which when executed by the one or more processors, cause the one or more processors to perform operations, the operations including
generating a plurality of trajectory candidates for an autonomous driving vehicle (ADV) from a starting point to an end point of a particular driving scenario;
generating a reference trajectory corresponding to the driving scenario based on a current state of the ADV associated with the starting point and an end state of the ADV associated with the end point, wherein the reference trajectory is associated with an objective;
for each of the trajectory candidates, comparing the trajectory candidate with the reference trajectory to generate an objective cost representing a similarity between the trajectory candidate and the reference trajectory; and
selecting one of the trajectory candidates as a target trajectory for driving the ADV based on objective costs of the trajectory candidates.

### Claim 16 (depends on 15)

The system of claim 15, wherein comparing the trajectory candidate with the reference trajectory comprises:
segmenting the trajectory candidate into a plurality of candidate segments;
segmenting the reference trajectory into a plurality of reference segments;
for each of the candidate segments, calculating a velocity difference between a velocity of the candidate segment and a velocity of a corresponding reference segment; and
calculating the objective cost based on the velocity differences between the candidate segments and reference segments.

### Claim 17 (depends on 16)

The system of claim 16, wherein the objective cost is calculated by summing squared velocity differences between the trajectory segments and the reference segments.

### Claim 18 (depends on 15)

The system of claim 15, further comprising:
for each of the trajectory candidates,
identifying an obstacle that is closest to the trajectory candidate,
measuring a distance between the trajectory candidate and the obstacle,
calculating a safety cost associated with the trajectory candidate based on the distance between the trajectory candidate and the obstacle, and
calculating a total trajectory cost for the trajectory candidate based on the objective cost and the safety cost; and
selecting the target trajectory from the trajectory candidates, wherein the target trajectory having a lowest total trajectory cost.

### Claim 19 (depends on 15)

The system of claim 15, further comprising:
for each of the trajectory candidates,
determining a changing rate of acceleration along the trajectory candidate,
calculating a comfort cost for the trajectory candidate based on the changing rate of acceleration along the trajectory candidate, and
calculating a total trajectory cost for the trajectory candidate based on the objective cost and the comfort cost; and
selecting the target trajectory from the trajectory candidates, wherein the target trajectory having a lowest total trajectory cost.

### Claim 20 (depends on 19)

The system of claim 19, wherein calculating a comfort cost for the trajectory candidate comprises:
segmenting the trajectory candidate into a plurality of candidate segments;
for each of the candidate segments, calculating a segment comfort cost for the candidate segment based on a changing rate of an acceleration associated with the candidate segment; and
calculating the comfort cost based on the segment comfort costs of the candidate segments.

### Claim 21 (depends on 20)

The system of claim 20, wherein the comfort cost is calculated based on a sum of segment comfort costs of the candidate segments.
