# US 2024/0004060 A1 · Automatic obstacle avoidance method, electronic device, and unmanned aerial vehicle

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

- Number: US 2024/0004060 A1
- Document ID: US20240004060A1
- Status: Published application
- Kind: A1
- Application number: 18/215,649
- Filed: 2023-06-28
- Published: 2024-01-04
- Earliest claimed: 2022-07-01
- Assignees: Autel Robotics Co., Ltd.
- Applicants: Autel Robotics Co., Ltd.
- Inventors: Feng TAO; Yanfeng DANG
- Web page: https://patentgrep.com/patent/US20240004060A1

## CPC

- G01S 13/935: for terrain-avoidance
- G01S 13/882: {for altimeters (measuring height using barometric means G01C5/06)}
- G05D 1/106: {Change initiated in response to external conditions, e.g. avoidance of elevated terrain or of no-fly zones}
- G05D 1/242: Means based on the reflection of waves generated by the vehicle (using passive navigation aids external to the vehicle G05D1/244; using signals provided by artificial sources external to the vehicle G05D1/247)
- G05D 1/622: Obstacle avoidance (predicting or avoiding probable or impending collision of road vehicles B60W30/08)
- G05D 2109/254: {Flying platforms, e.g. multicopters}
- G05D 2111/30: Radio signals
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES

## Abstract

An automatic obstacle avoidance method is applied to an unmanned aerial vehicle (UAV). The UAV includes at least one radar. The method includes: acquiring detected data of the radar; calculating theoretical state information of the detection target based on motion information of the UAV; excluding invalid data from the detected data using a Kalman filtering algorithm to obtain corrected data by combining the detected data with the theoretical state information; determining the height information of the UAV by correcting the detected data; and triggering an obstacle avoidance warning when the height information of the UAV is less than a pre-set height threshold value.

## Claims

### Claim 1

An automatic obstacle avoidance method for a carrier, comprising:
acquiring detected data of a radar of the carrier, the detected data comprising observation state information of a detection target;
calculating theoretical state information of the detection target based on motion information of the carrier;
excluding invalid data from the detected data to obtain corrected data by combining the detected data with the theoretical state information;
determining height information of the carrier according to the corrected data; and
triggering an obstacle avoidance warning when the height information of the carrier is less than a first pre-set height threshold value.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the invalid data comprises observation state information of a non-ground target, the non-ground target is a detection target having an elapsed time of less than a second pre-set threshold value, and the elapsed time is a time required for the carrier to pass the detection target.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the excluding invalid data from the detected data to obtain corrected data by combining the detected data with the theoretical state information comprises:
iterating a plurality of Kalman filtering processes based on the observation state information and the theoretical state information; and
outputting state information estimate of the detection target after each Kalman filtering process as the corrected data.

### Claim 4 (depends on 3)

The method according to claim 3, wherein the invalid data further comprises:
observation state information of false detection targets formed by external noise.

### Claim 5 (depends on 3)

The method according to claim 3, wherein the Kalman filtering process comprises:
calculating a prior estimate value of the theoretical state information through a pre-set first transformation relation based on the state information estimate obtained from a previous Kalman filtering process;
calculating a current state vector estimation error covariance matrix based on a state vector estimation error covariance matrix obtained from the previous Kalman filtering process;
calculating a residual between the prior estimate value and the observation state information according to a pre-set second transformation relation;
calculating a Kalman gain coefficient according to the current state vector estimation error covariance matrix, the prior estimate value, and the observation state information;
updating the current state vector estimation error covariance matrix according to the Kalman gain coefficient; and
weighting and superimposing the residual and the prior estimate value with the Kalman gain coefficient as a weighting coefficient to obtain the state information estimate.

### Claim 6 (depends on 5)

The method according to claim 5, wherein the theoretical state information comprises anyone or any combination of: position information, velocity information, and acceleration information of the detection target in a three-dimensional coordinate system; and
the observation state information comprises: a distance between the detection target and the radar, a direction angle of the detection target with respect to the radar, a pitch angle of the detection target with respect to the radar, and a radial speed of the detection target with respect to the radar.

### Claim 7 (depends on 5)

The method according to claim 5, wherein the pre-set first transformation relation is established based on an acceleration model of the carrier, the pre-set second transformation relation is a corresponding relation between the theoretical state information and the observation state information after linearization.

### Claim 8 (depends on 7)

The method according to claim 7, further comprising:
correcting the prior estimate value of the theoretical state information by acceleration information of the carrier.

### Claim 9 (depends on 1)

The method according to claim 1, wherein the determining the height information of the carrier according to the corrected data comprises:
determining a tilt angle of the radar according to posture information of the carrier;
determining whether there is a detection target within a search range in the corrected data, the search range being formed by floating up and down an allowable error based on the tilt angle;
in response to determining that there is a detection target within the search range, selecting a detection target closest to the tilt angle as a reference target within the search range;
calculating current height information of the carrier according to the observation state information of the reference target; and
in response to determining that there is no detection target within the search range, estimating the current height information of the carrier according to motion information of the carrier and previously calculated height information.

### Claim 10 (depends on 9)

The method according to claim 9, wherein the radar is disposed at a bottom of the carrier proximate to the ground;
the determining a tilt angle of the radar according to posture information of the carrier comprises:
acquiring a pitch angle of the carrier as an azimuth angle of the radar, and
acquiring a roll angle of the carrier as a pitch angle of the radar.

### Claim 11 (depends on 10)

The method according to claim 10, further comprising:
acquiring quaternions of the carrier; and
converting the quaternions to Euler angles to acquire the pitch angle of the carrier and the roll angle of the carrier.

### Claim 12 (depends on 10)

The method according to claim 10, wherein the calculating current height information of the carrier according to the observation state information of the reference target comprises:
acquiring a distance between the reference target and the radar, and an azimuth angle of the radar;
calculating current height information of the carrier based on the distance between the reference target and the radar and the azimuth angle by a trigonometric function.

### Claim 13 (depends on 12)

The method according to claim 12, wherein the current height information of the carrier is calculated by using the following formula:
H=R cos β
wherein H is the current height information of the carrier, R is the distance between the reference target and the radar, and β is the azimuth angle of the radar.

### Claim 14 (depends on 10)

The method according to claim 10, wherein the current height information of the carrier is estimated by using the following formula:
H(k)=H(k−1)+vt+0.5at2
wherein H(k) is the current height information of the carrier, and H(k−1) is a previously calculated height information; v is a velocity of the carrier in a height direction; a is an acceleration of the carrier in the height direction; t is time elapsed between H(k) and H(k−1).

### Claim 15

An electronic device, comprising at least one processor, and a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the following operations:
acquiring detected data of a radar of a carrier, the detected data comprising observation state information of a detection target;
calculating theoretical state information of the detection target based on motion information of the carrier;
excluding invalid data from the detected data to obtain corrected data by combining the detected data with the theoretical state information;
determining height information of the carrier according to the corrected data; and
triggering an obstacle avoidance warning when the height information of the carrier is less than a first pre-set height threshold value.

### Claim 16

An unmanned aerial vehicle (UAV), comprising:
a vehicle body;
a radar disposed at the bottom of the vehicle body proximate to the ground;
a vehicle arm coupled to the vehicle body;
a power drive unit provided on the vehicle arm for providing flying power to the UAV; and
a flight controller provided on the vehicle body and communicatively coupled to the radar; wherein the flight controller is configured to:
acquire detected data of the radar, the detected data comprising observation state information of a detection target;
calculate theoretical state information of the detection target based on motion information of the UAV;
exclude invalid data from the detected data to obtain corrected data by combining the detected data with the theoretical state information; and
determine height information of the UAV according to the corrected data;
trigger an obstacle avoidance warning when the height information of the UAV is less than a first pre-set height threshold value.

### Claim 17 (depends on 16)

The UAV according to claim 16, wherein the invalid data comprises observation state information of a non-ground target, the non-ground target is a detection target having an elapsed time of less than a second pre-set threshold value, and the elapsed time is a time required for the UAV to pass the detection target.

### Claim 18 (depends on 16)

The UAV according to claim 16, wherein the flight controller is further configured to:
iterate a plurality of Kalman filtering processes based on the observation state information and the theoretical state information; and
output state information estimate of the detection target after each Kalman filtering process as the corrected data.

### Claim 19 (depends on 18)

The UAV according to claim 18, wherein the Kalman filtering process comprises:
calculating a prior estimate value of the theoretical state information through a preset first transformation relation based on the state information estimate obtained from a previous Kalman filtering process;
calculating a current state vector estimation error covariance matrix based on a state vector estimation error covariance matrix obtained from the previous Kalman filtering processing;
calculating a residual between the prior estimate value and the observation state information according to a pre-set second transformation relation;
calculating a Kalman gain coefficient according to the current state vector estimation error covariance matrix, the prior estimate value, and the observation state information;
updating the current state vector estimation error covariance matrix according to the Kalman gain coefficient; and
weighting and superimposing the residual and the prior estimate value with the Kalman gain coefficient as a weighting coefficient to obtain the state information estimate.

### Claim 20 (depends on 16)

The UAV according to claim 16, wherein the flight controller is further configured to:
determine a tilt angle of the radar according to posture information of the UAV;
determine whether there is a detection target within a search range in the corrected data, the search range being formed by floating up and down an allowable error based on the tilt angle;
in response to determining that there is a detection target within the search range, select a detection target closest to the tilt angle as a reference target within the search range;
calculate current height information of the UAV according to the observation state information of the reference target; and
in response to determining that there is no detection target within the search range, estimate the current height information of the UAV according to motion information of the UAV and previously calculated height information.
