CN-122018511-A - Follow-up control method and device of mobile equipment, electronic equipment and storage medium

CN122018511ACN 122018511 ACN122018511 ACN 122018511ACN-122018511-A

Abstract

The invention provides a following control method, a device, electronic equipment and a storage medium of mobile equipment, which are used for maintaining target stability by judging consistency of candidate targets in perception data, realizing target confirmation and locking when an interactive trigger instruction corresponding to the candidate targets is detected, continuously outputting target area information and fusing depth/distance information to calculate relative pose in a locking state, and converting the relative pose into distance errors and angle errors to generate a chassis control instruction so as to realize closed-loop adjustment of a following process. Therefore, the false following probability in a multi-person scene can be reduced, the stability and controllability of target locking and following are improved, and smoother and more accurate following effects are maintained when targets move and environments change.

Inventors

JIANG ZHUMIAO
CHEN ZHAO
LU XIAOSONG
Pu tianfa

Assignees

宁波视睿迪光电有限公司

Dates

Publication Date: 20260512
Application Date: 20260326

Claims (10)

1. A following control method of a mobile device, characterized by comprising: acquiring personnel related perception data in a perception range of mobile equipment, detecting personnel candidate targets according to the personnel related perception data, and carrying out consistency judgment on the candidate targets; when an interaction triggering instruction corresponding to the candidate target is detected, determining that the candidate target is a following target, and entering a locking state; Under the locking state, continuously tracking the following target to obtain target area information of the following target, and calculating relative pose information of the following target relative to the mobile device according to the target area information; And generating a chassis motion control instruction according to the relative pose information and transmitting the chassis motion control instruction to the mobile chassis so as to drive the mobile chassis to follow the following target.
2. The method according to claim 1, wherein acquiring person-related awareness data within a awareness range of a mobile device, detecting person candidate targets based on the person-related awareness data, and performing consistency determination on the candidate targets, specifically comprises: Collecting image data as related perception data of the personnel through a binocular camera; Calculating depth and distance information according to binocular parallax, and storing the depth and distance information in association with the image data; performing portrait identification for the person-related awareness data to output the candidate target; The consistency determination is performed to maintain target consistency based on feature vectors, apparent consistency, and temporal continuity of the candidate targets.
3. The method according to claim 1, wherein when an interaction trigger instruction corresponding to the candidate target is detected, determining that the candidate target is a following target and entering a locked state, specifically comprises: Receiving at least one of a gesture instruction, a voice instruction, a touch instruction and a selection instruction issued by a mobile terminal application as the interaction triggering instruction; when the interaction triggering instruction is successfully associated with the candidate target, determining the candidate target as the following target and entering a locking state; Invoking a preset gesture model library to identify gestures of the candidate targets and outputting gesture types; and when the gesture category belongs to a preset set, generating the interaction triggering instruction, wherein the preset set at least comprises one or more of palm, fist making, scissor hand and Okey gestures.
4. The method according to claim 1, wherein in the locked state, continuous tracking is performed on the following target, target area information of the following target is obtained, and relative pose information of the following target relative to the mobile device is calculated according to the target area information, specifically including: Generating a target bounding box or a target contour representing a following target in real time in a locked state; updating the target bounding box or target contour with the following target movement to form the target region information; Performing continuous tracking on the following target by adopting a filtering target tracking algorithm and outputting the target area information; calculating pixel coordinates and variation thereof according to the target area information; And fusing depth and distance information and the pixel coordinates, and calculating the relative pose information, wherein the relative pose information comprises at least one of a target relative distance and a target relative azimuth angle.
5. The method according to claim 1, wherein generating chassis motion control instructions according to the relative pose information and issuing the chassis motion control instructions to a mobile chassis to drive the mobile chassis to follow the following target, specifically comprising: Converting the relative pose information into a distance error and an angle error; generating the chassis motion control instruction based on the distance error and the angle error; and issuing the chassis motion control instruction to control at least one of moving speed, acceleration, steering angular speed, moving distance and steering angle, so that the error is converged within a threshold range.
6. The method according to claim 1, wherein the method further comprises: detecting a start following interaction instruction and a stop following interaction instruction in a locking state; Issuing an intelligent follow-up starting instruction to the mobile chassis when the follow-up starting interaction instruction is detected; and when the following stopping interaction instruction is detected, an intelligent following closing instruction is issued to the mobile chassis, and the following is stopped.
7. The method according to claim 1, characterized in that: mapping the palm gesture recognition result into a start following interaction instruction; And mapping the recognition result of the fist-making gesture to stop following the interactive instruction.
8. A following control device of a mobile apparatus, characterized by comprising: the target determining module is used for acquiring personnel related perception data in a perception range of the mobile equipment, detecting personnel candidate targets according to the personnel related perception data, and carrying out consistency judgment on the candidate targets; the locking triggering module is used for determining that the candidate target is a following target and entering a locking state when an interaction triggering instruction corresponding to the candidate target is detected; The pose determining module is used for continuously tracking the following target in the locking state to obtain target area information of the following target, and calculating relative pose information of the following target relative to the mobile device according to the target area information; and the movement control module is used for generating a chassis movement control instruction according to the relative pose information and sending the chassis movement control instruction to the movement chassis so as to drive the movement chassis to follow the following target.
9. An electronic device comprising a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication over the bus when the electronic device is in operation, the machine-readable instructions when executed by the processor performing the steps of the following control method of a mobile device as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of the following control method of a mobile device according to any one of claims 1 to 7.

Description

Follow-up control method and device of mobile equipment, electronic equipment and storage medium Technical Field The disclosure relates to the technical field of computer vision control, in particular to a following control method and device of mobile equipment, electronic equipment and a storage medium. Background With the increase of applications of service robots, accompanying robots and intelligent terminals with mobility in scenes such as families, commercial exhibition, navigation explanation and the like, mobile equipment automatically follows according to personnel position changes, and the intelligent terminal has become an important capability for improving interaction experience and service efficiency. The existing following scheme generally relies on a camera, a laser radar or an ultrasonic sensor to acquire environmental information, estimates the relative position of personnel through a target detection/tracking algorithm, and further controls the chassis to move so as to realize following. However, the prior art still has the defects that on one hand, under the conditions of crowd interference, shielding, illumination change or rapid movement of a target, the mobile equipment easily causes the problems of target following drift, target following mistake or target loss, and particularly when a plurality of candidate persons appear in a perception range, the system is difficult to continuously maintain the following relation of the same person. On the other hand, many schemes lack natural, low-cost and interpretable man-machine interaction triggering modes, control such as confirmation, starting, stopping and the like following targets often depends on preset fixed processes or complex peripheral operations, so that target locking is not visual, false triggering probability is high, and requirements of quick switching, instant control and safety and controllability in a real scene are difficult to meet. Disclosure of Invention The embodiment of the disclosure provides at least a following control method, a device, an electronic device and a storage medium of mobile equipment, wherein the following control method, the device, the electronic device and the storage medium are used for keeping stability of a target by judging consistency of a candidate target in perception data, realizing target confirmation and locking when an interactive trigger instruction corresponding to the candidate target is detected, continuously outputting target area information in a locking state, fusing depth/distance information to calculate a relative pose, and converting the relative pose into a distance error and an angle error to generate a chassis control instruction, so that closed-loop adjustment of a following process is realized. Therefore, the false following probability in a multi-person scene can be reduced, the stability and controllability of target locking and following are improved, and smoother and more accurate following effects are maintained when targets move and environments change. The embodiment of the disclosure provides a following control method of mobile equipment, which comprises the following steps: acquiring personnel related perception data in a perception range of mobile equipment, detecting personnel candidate targets according to the personnel related perception data, and carrying out consistency judgment on the candidate targets; when an interaction triggering instruction corresponding to the candidate target is detected, determining that the candidate target is a following target, and entering a locking state; Under the locking state, continuously tracking the following target to obtain target area information of the following target, and calculating relative pose information of the following target relative to the mobile device according to the target area information; And generating a chassis motion control instruction according to the relative pose information and transmitting the chassis motion control instruction to the mobile chassis so as to drive the mobile chassis to follow the following target. In an optional implementation manner, acquiring personnel related perception data in a perception range of a mobile device, detecting personnel candidate targets according to the personnel related perception data, and performing consistency judgment on the candidate targets, wherein the method specifically comprises the following steps: Collecting image data as related perception data of the personnel through a binocular camera; Calculating depth and distance information according to binocular parallax, and storing the depth and distance information in association with the image data; performing portrait identification for the person-related awareness data to output the candidate target; The consistency determination is performed to maintain target consistency based on feature vectors, apparent consistency, and temporal continuity of the candidate targets. In an alternative embodime