Methods, systems, device, and storage mediums for obstacle avoidance of surgical robots

1 . A method for obstacle avoidance of a surgical robot implemented by a system including a first acquisition device, a second acquisition device, a processor, and the surgical robot, the method comprising:

causing the first acquisition device to collect first data, wherein the first data is image data of a space in which a target subject is located;

causing the processor to construct a safety zone for the target subject based on the first data;

causing the second acquisition device to collect second data outside the safety zone, wherein the second data is image data of the target subject, the second acquisition device is carried by a mechanical arm of the surgical robot to collect the second data, and the second acquisition device and the mechanical arm are not allowed to move into the safety zone during the process of collecting the second data; and

causing the processor to construct a three-dimensional model of the target subject based on the second data; and

causing the processor to perform obstacle avoidance detection based on the three-dimensional model of the target subject during an operation of the surgical robot.

2 . The method of claim 1 , wherein the causing the processor to construct a safety zone for the target subject based on the first data includes:

causing the processor to identify first subset data of the target subject based on the first data;

causing the processor to determine a center position of the target subject based on the first subset data; and

causing the processor to determine a spherical region as the safety zone based on the center position and a preset radius.

3 . The method of claim 1 , wherein the causing the second acquisition device to collect second data outside the safety zone includes:

causing the processor to determine a plurality of first acquisition points based on the safety zone, the plurality of first acquisition points being positions outside the safety zone where the second acquisition device collects the second data; and

causing the second acquisition device to sequentially collect the second data at the plurality of first acquisition points.

4 . The method of claim 3 , wherein the causing the processor to determine a plurality of first acquisition points based on the safety zone includes:

causing the processor to divide the safety zone into a plurality of regions; and

causing the processor to generate the plurality of first acquisition points around a periphery of the plurality of regions based on a preset generation algorithm.

5 . The method of claim 3 , wherein the causing the second acquisition device to sequentially collect the second data at the plurality of first acquisition points includes:

for any first acquisition point of the plurality of first acquisition points,

causing the processor to determine whether a coverage rate of second data collected at the first acquisition point and previous first acquisition points relative to the target subject satisfies a requirement;

in response to determining that the coverage rate does not satisfy the requirement, causing the second acquisition device to collect second data at a next first acquisition point; and

in response to determining that the coverage rate satisfies the requirement, causing the processor to end the collection.

6 . The method of claim 5 , wherein the coverage rate is a ratio of an area of the safety zone covered by the second data collected at the first acquisition point and the previous first acquisition points to a total surface area of the safety zone, and the requirement includes that the coverage rate is greater than a preset minimum coverage rate threshold.

7 . The method of claim 3 , further comprising:

causing the processor to determine whether the collected second data includes image data of a preset location of the target subject;

in response to determining that the collected second data does not include the image data of the preset location of the target subject, causing the processor to determine a second acquisition point, and causing the second acquisition device to collect the image data of the preset position of the target subject at the second acquisition point.

8 . The method of claim 1 , wherein the second acquisition device is located at an end of the mechanical arm of the surgical robot.

9 . The method of claim 1 , further comprising:

causing the processor to construct an initial three-dimensional model of the space in which the target subject is located based on the first data;

causing the processor to determine a three-dimensional model of the space in which the target subject is located based on the three-dimensional model of the target subject and the initial three-dimensional model of the space, and

causing the processor to perform the obstacle avoidance detection based on the three-dimensional model of the space.

10 . The method of claim 9 , wherein the causing the processor to construct an initial three-dimensional model of the space in which the target subject is located based on the first data includes:

causing the first acquisition device to update the first data in real time; and

causing the processor to construct the initial three-dimensional model of the space in which the target subject is located based on the updated first data.

11 . The method of claim 9 , wherein the method further includes:

causing the processor to divide the three-dimensional model of the space into a plurality of regions;

causing the processor to determine a simulation completeness degree of three-dimensional models constructed for each region of the plurality of regions;

causing the processor to determine a required simulation completeness degree for each region based on a distance of the region to the target subject and an activity frequency of the surgical robot in the region; and

causing the first acquisition device to adjust a collection process of the first acquisition device based on the simulation completeness degree and the required simulation completeness degree of each region.

12 . The method of claim 9 , wherein the causing the processor to construct an initial three-dimensional model of the space in which the target subject is located based on the first data includes:

causing the processor to determine a relative position of the target subject and the first acquisition device based on the first data;

causing the first acquisition device to adjust a shooting angle of the first acquisition device based on the relative position;

causing the first acquisition device to obtain new image data at the adjusted shooting angle; and

causing the first acquisition device to designate the new image data as the first data for constructing the initial three-dimensional model.

13 . The method of claim 12 , wherein the causing the first acquisition device to adjust a shooting angle of the first acquisition device based on the relative position includes:

causing the processor to determine an adjusted angle based on the relative position;

causing the processor to determine, through a correction value determination model, a correction value of the adjusted angle based on a type of surgery, participant information, and the adjusted angle, the correction value determination model being a deep learning neural network model; and

causing the first acquisition device to adjust the shooting angle of the first acquisition device based on the correction value of the adjusted angle.

14 . The method of claim 1 , wherein the first acquisition device and the second acquisition device are integrated into a same device.

15 . The method of claim 14 , wherein one of the first acquisition device and the second acquisition device is canceled, and functions of the first acquisition device and the second acquisition device are implemented by the other of the first acquisition device and the second acquisition device.

16 . The method of claim 2 , wherein the preset radius is an adaptively generated safety zone radius determined by the processor by:

constructing three-dimensional models of a head of the target subject based on the first data and head data collected before surgery;

determining a similarity of the three-dimensional models of the head as a confidence level of the first data; and

determining the preset radius based on the confidence level, the preset radius being negatively correlated with the confidence level.

17 . The method of claim 1 , wherein a size of the safety zone is determined by the processor based on a count of historical actual acquisition points of a similar target subject of the target subject, the count of historical actual acquisition points is a count of data acquisition points collected by the second acquisition device in historical data when a data coverage rate corresponding to the similar target subject satisfies a requirement.

18 . The method of claim 9 , wherein:

the causing the processor to determine a three-dimensional model of the space includes:

causing the processor to load the three-dimensional model of the target subject into the initial three-dimensional model of the space; and

causing the processor to determine the three-dimensional model of the space by removing the safety zone of the target subject from the initial three-dimensional model loaded with the three-dimensional model of the target subject,

the causing the processor to perform the obstacle avoidance detection based on the three-dimensional model of the space includes: the causing the processor to perform path planning for the mechanical arm of the surgical robot in the three-dimensional model of the space.

19 . A system for obstacle avoidance of a surgical robot, comprising a first acquisition device, a second acquisition device, the surgical robot, and a processor, wherein

the first acquisition device is configured to collect first data, wherein the first data is image data of a space in which a target subject is located;

the second acquisition device is configured to collect second data outside a safety zone of the target subject, wherein the second data is image data of the target subject, the second acquisition device is carried by a mechanical arm of the surgical robot to collect the second data, and the second acquisition device and the mechanical arm are not allowed to move into the safety zone during the process of collecting the second data;

the surgical robot is configured to perform a surgical procedure;

the processor is configured to:

construct the safety zone based on the first data; and

construct a three-dimensional model of the target subject based on the second data; and

perform obstacle avoidance detection based on the three-dimensional model of the target subject during an operation of the surgical robot.

20 . A computer-readable storage medium storing computer instructions, wherein when reading the computer instructions from the storage medium, a computer implements the method for obstacle avoidance of a surgical robot, wherein the method comprises:

causing a first acquisition device to collect first data, wherein the first data is image data of a space in which a target subject is located;

causing the computer to construct a safety zone for the target subject based on the first data;

causing a second acquisition device to collect second data outside the safety zone, wherein the second data is image data of the target subject, the second acquisition device is carried by a mechanical arm of the surgical robot to collect the second data, and the second acquisition device and the mechanical arm are not allowed to move into the safety zone during the process of collecting the second data; and

causing the computer to construct a three-dimensional model of the target subject based on the second data; and

causing the computer to construct an initial three-dimensional model of the space in which the target subject is located based on the first data;

causing the computer to determine a three-dimensional model of the space in which the target subject is located based on the three-dimensional model of the target subject and the initial three-dimensional model of the space, and

causing the computer to perform obstacle avoidance detection based on the three-dimensional model of the space during an operation of the surgical robot.