Method and device for detecting obstacle speed, computer device, and storage medium

Patent metadata

Field	Value
Publication number	US-11226350-B2
Application number	US-201916562043-A
Country	US
Kind code	B2
Filing date	Sep 5, 2019
Priority date	Sep 7, 2018
Publication date	Jan 18, 2022
Grant date	Jan 18, 2022

Abstract

Embodiments of the present disclosure provide a method and device for detecting a speed of an obstacle, a computer device, and a storage medium. The method includes: calculating at least two real-time speeds corresponding to the obstacle by using a multi-frame difference algorithm according to multi-frame data acquired by a sensor in a preset time window; calculating at least two speed statistic values corresponding to the obstacle according to the at least two real-time speeds; mapping each of the at least two speed statistic values to a corresponding static probability according to a mapping relationship between speed statistic values and static probabilities, to obtain at least two static probabilities; and fusing the at least two static probabilities to obtain a fused static probability of the obstacle, and determining the speed of the obstacle according to the fused static probability.

First claim

What is claimed is:

1. A method for detecting a speed of an obstacle, comprising:
calculating at least two real-time speeds corresponding to the obstacle by using a multi-frame difference algorithm according to multi-frame data acquired by a sensor in a preset time window;
calculating at least two speed statistic values corresponding to the obstacle according to the at least two real-time speeds;
mapping each of the at least two speed statistic values to a corresponding static probability according to a mapping relationship between speed statistic values and static probabilities, to obtain at least two static probabilities; and
fusing the at least two static probabilities to obtain a fused static probability of the obstacle, and determining the speed of the obstacle according to the fused static probability.

2. The method according to claim 1 , wherein the at least two speed statistic values comprise: a variance of speed modulus, a mean value of speed angle difference, and a second-order variance of the speed modulus.

3. The method according to claim 2 , wherein fusing the at least two static probabilities to obtain the fused static probability of the obstacle comprises:
calculating the fused static probability P final of the obstacle by:

P
final

=

e

-

∑

logit
⁡

(

P
i

)

1
+

e

-

∑

logit
⁡

(

P
i

)

,

i
∈

[

2
,
N

]

,

where
⁢

⁢

logit
⁡

(

P
i

)

=

log
⁢

P
i

1
-

P
i

,

N denotes a number of the static probabilities, and P i denotes the static probability corresponding to the i th speed statistic value.

4. The method according to claim 1 , wherein the mapping relationship between speed statistic values and static probabilities comprises:

P
=

{

0
,

v
<
t

1
-

e

-

(

v
-
t

s

)

2

,

v
≥
t

where v denotes the speed statistic value, P denotes the static probability, and t and s denote preset mapping parameters.

5. The method according to claim 4 , wherein different mapping parameters are assigned to different speed statistic values;
wherein the preset mapping parameter t corresponding to the variance of the speed modulus is greater than the preset mapping parameter t corresponding to the mean value of the speed angle difference, and the preset mapping parameter t corresponding to the mean value of the speed angle difference is greater than the preset mapping parameter t corresponding to the second-order variance of the speed modulus;
the speed statistic value v corresponding to the variance of speed modulus is greater than the speed statistic value v corresponding to the mean value of the speed angle difference, and the speed statistic value v corresponding to the mean value of the speed angle difference is greater than the speed statistic value v corresponding to the second-order variance of the speed modulus.

6. The method according to claim 1 , wherein determining the speed of the obstacle according to the fused static probability comprises:
comparing the fused static probability with a probability threshold;
determining that the obstacle is in a static state in response to determining that the fused static probability is greater than or equal to the probability threshold; and
determining the real-time speed as the speed of the obstacle in response to determining that the fused static probability is less than the probability threshold.

7. A device for detecting a speed of an obstacle, comprising:
one or more processors;
a storage device, configured to store one or more programs,
wh

Assignees

• Baidu online network technology beijing co ltd
• Apollo Intelligent Driving Tech Beijing Co Ltd

Inventors

• Cheng Kai

Classifications

• G06V20/58 Primary

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title

• G01S13/58 Primary

  Velocity or trajectory determination systems; Sense-of-movement determination systems · CPC title

• G06V20/10

  Terrestrial scenes (scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59) · CPC title

• G01S17/58

  Velocity or trajectory determination systems; Sense-of-movement determination systems · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title