CN114986524A - Robotic arm trajectory planning method, device and electronic device based on brain-computer interface - Google Patents

Abstract

The invention provides a method, device and electronic device for trajectory planning of a robotic arm based on a brain-computer interface, and relates to the field of robotics. The method includes: acquiring a first configuration list; performing collision detection based on a three-dimensional head area and the first configuration list , and delete the 3D point in the first configuration list where the robot collides with the 3D head area and the robot arm joint angle data corresponding to the 3D point based on the collision detection result; take the deleted first configuration list as the second configuration list, and obtain the target access sequence of the robotic arm to all 3D points in the second configuration list; based on the second configuration list and the target access sequence, obtain the simulated implantation trajectory formed by the movement of the robotic arm in the 3D head area, and simulate the implantation trajectory It is used to implant the brain-computer interface into the three-dimensional head area, which solves the problem that the simulated implantation trajectory formed by the movement of the robotic arm in the highly cluttered three-dimensional head area cannot be quickly and safely obtained.

Description

Robotic arm trajectory planning method, device and electronic device based on brain-computer interface

technical field

The invention relates to the field of robotics technology, and in particular, to a method, device and electronic device for trajectory planning of a robotic arm based on a brain-computer interface.

Background technique

Brain-computer interface (BCI) refers to the direct connection created between the human or animal brain and external devices to realize the exchange of information between the brain and the device. The brain-computer interface collects EEG signals from the cerebral cortex through signal acquisition equipment, and converts the EEG signals into target signals that can be recognized by the computer through operations such as amplification, filtering, and A/D conversion, and then preprocesses the target signals. The feature signal is extracted from the preprocessed target signal, and then the feature signal is used for pattern recognition to realize the information interaction between the brain and the device.

In the prior art, because the head area is a highly cluttered three-dimensional environment, the robotic arm is prone to collide with the three-dimensional head area during the process of implanting the brain-computer interface into the three-dimensional head area, which not only increases the number of implants It will also cause many production safety problems, making it impossible to quickly and safely obtain the simulated implantation trajectory formed by the movement of the robotic arm in the highly cluttered three-dimensional head area, and then based on the simulated implantation trajectory The interface is implanted into the three-dimensional head region.

Therefore, in the prior art, the simulated implantation trajectory formed by the movement of the robotic arm in the highly cluttered 3D head region cannot be quickly and safely obtained, and then the technology of implanting the brain-computer interface into the 3D head region based on the simulated implantation trajectory There is no effective solution to the problem for those skilled in the related art.

SUMMARY OF THE INVENTION

The present invention provides a method, device and electronic device for trajectory planning of a robotic arm based on a brain-computer interface, which are used to solve the problem of the inability to quickly and safely obtain the simulated implant formed by the movement of a robotic arm in a highly cluttered three-dimensional head region in the prior art. Then, based on the technical problem of implanting the brain-computer interface into the three-dimensional head area based on the simulated implantation trajectory, the implantation efficiency and safety of the brain-computer interface are improved, and many production safety problems are avoided.

The present invention provides a method for planning a trajectory of a robotic arm based on a brain-computer interface, comprising: acquiring a first configuration list, where the first configuration list includes each three-dimensional point in a three-dimensional head area and the position of the robotic arm in the three-dimensional head area. The joint angle data of the manipulator at each three-dimensional point in the area; perform collision detection based on the three-dimensional head area and the first configuration list, and delete the manipulator and the three-dimensional manipulator in the first configuration list based on the collision detection result The 3D points that collide in the head area and the joint angle data of the robot arm corresponding to the 3D points; take the deleted first configuration list as the second configuration list, and obtain all the 3D points of the robot arm in the second configuration list The target access sequence of Implanted into the 3D head region.

According to a method for planning a trajectory of a manipulator based on a brain-computer interface provided by the present invention, the acquiring the first configuration list includes: acquiring pose data of the manipulator at each three-dimensional point in the three-dimensional head region of the manipulator, so The manipulator pose data includes the three-dimensional coordinate value of the manipulator and the rotation angle of the coordinate axis; the manipulator pose data is converted into the manipulator joint angle data based on the kinematic inverse solution operation strategy, and the manipulator joint angle The data includes a plurality of manipulator joints and the rotation angle of each manipulator joint; the first configuration list is obtained based on each three-dimensional point in the three-dimensional head region and the manipulator joint angle data corresponding to each three-dimensional point.

According to a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention, the collision detection is performed based on the three-dimensional head region and the first configuration list, and the first configuration list is deleted based on the collision detection result. The 3D points where the robot arm collides with the 3D head area and the joint angle data of the robot arm corresponding to the 3D points include: obtaining the first bounding box of the robot arm corresponding to each 3D point in the first configuration list, and a second bounding box corresponding to the three-dimensional head area; for each of the first bounding boxes, determine whether the first bounding box intersects the second bounding box; When the second bounding box intersects, determine whether the robotic arm intersects the three-dimensional head area at the three-dimensional point corresponding to the first bounding box; When the three-dimensional head regions intersect, the three-dimensional point and its corresponding mechanical arm joint rotation angle data are deleted from the first configuration list.

According to a method for trajectory planning of a robotic arm based on a brain-computer interface provided by the present invention, the acquisition of the first bounding box corresponding to each three-dimensional point of the robotic arm at each three-dimensional point in the first configuration list and the three-dimensional head region The corresponding second bounding box includes: obtaining the coordinate extreme value data of the manipulator at each three-dimensional point in the first configuration list, where the coordinate extreme value data of the manipulator includes a plurality of manipulator arms of the manipulator. The coordinate extreme value data of the joint and each manipulator joint; the first coordinate corresponding to each manipulator joint in the manipulator at each three-dimensional point in the first configuration list is obtained based on the coordinate extreme value data of the manipulator A bounding box; obtaining the regional coordinate extreme value data of the three-dimensional head region, and obtaining a second bounding box corresponding to the three-dimensional head region based on the regional coordinate extreme value data.

According to a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention, the acquiring the target access sequence of the robotic arm to all the three-dimensional points in the second configuration list includes: acquiring the current iteration count and the the current access order of all three-dimensional points in the second configuration list; based on the current access order and the three-dimensional coordinates of each three-dimensional point in the second configuration list, obtain the current access path corresponding to the current access order; Obtain the local access sequence between any two non-adjacent three-dimensional points in the current access sequence, and perform inversion processing on the local access sequence in the current access sequence to obtain a first access sequence; order and the three-dimensional coordinates of each three-dimensional point in the second configuration list, obtain the first access path corresponding to the first access sequence; update the current access path based on the current access path and the first access path sequence and the current iteration count; take the updated current access sequence as the new current access sequence, and use the updated current iteration count as the new current iteration count; repeat the above steps until the current iteration count reaches the maximum iteration count number of times, and the final current access sequence is taken as the target access sequence.

According to a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention, the simulated implant formed by the movement of the robotic arm in the three-dimensional head region is obtained based on the second configuration list and the target access sequence. entering the trajectory, including: constructing a 3D point cascade diagram based on the target access sequence and the second configuration list; the 3D point cascade diagram includes a plurality of cascaded 3D point layers and multiple 3D point layers corresponding to each 3D point layer. vertices, where each 3D point layer corresponds to a 3D point, and multiple vertices represent multiple sets of mechanical arm joint angle data corresponding to the 3D point; traverse each 3D point layer in the 3D point cascade diagram, in the adjacent One vertex is selected from each of the two 3D point layers, and based on the selected two vertices, the movement path and movement cost of the robotic arm moving between the two adjacent 3D point layers are obtained; All vertices in the three-dimensional point cascade graph, and finally obtain all the moving paths and the moving cost; traverse each three-dimensional point layer in the three-dimensional point cascade graph, based on the moving cost from two adjacent three-dimensional points Screening out target movement paths from the multiple movement paths corresponding to the point layer to obtain multiple target movement paths; obtaining a target configuration list based on the multiple target movement paths and the second configuration list, and obtaining a robotic arm based on the target configuration list A simulated implant trajectory formed in the 3D head region.

According to a method for planning the trajectory of a robotic arm based on a brain-computer interface provided by the present invention, the moving path and the moving cost of the robotic arm moving between the two adjacent three-dimensional point layers are obtained based on the two selected vertices, including: : Obtain the manipulator joint angle data corresponding to the two selected vertices and the maximum angle angular velocity of the manipulator joint; obtain the manipulator joint angle difference based on the manipulator joint angle data corresponding to the two vertices, and based on the manipulator arm Obtain the maximum joint corner time based on the joint corner difference and the maximum corner angular velocity; obtain the corner symbol switching cost of the manipulator joint based on the manipulator joint corner data corresponding to the two vertices; based on the corner symbol switching cost and the preset additional cost coefficient, to obtain the extra cost of the manipulator moving between the two adjacent three-dimensional point layers; based on the maximum joint rotation time and the extra cost, to obtain the manipulator arm to move between the two adjacent three-dimensional point layers Mobile cost of moving.

The present invention also provides a robotic arm trajectory planning device based on a brain-computer interface, comprising: a configuration acquisition module for acquiring a first configuration list, where the first configuration list includes each three-dimensional point in the three-dimensional head region and the mechanical Robotic arm joint angle data at each 3D point of the arm in the 3D head area; a collision detection module for performing collision detection based on the 3D head area and the first configuration list, and based on the collision detection As a result, delete the 3D point in the first configuration list where the robot arm collides with the 3D head area and the robot arm joint rotation angle data corresponding to the 3D point; the sequence determination module is used to use the deleted first configuration list as the second configuration list. a configuration list, and obtain the target access order of the robotic arm to all the three-dimensional points in the second configuration list; the trajectory generation module is configured to obtain the robot arm in the second configuration list and the target access order based on the second configuration list. A simulated implant trajectory formed by motion in a three-dimensional head region, the simulated implant trajectory being used to implant a brain-computer interface into the three-dimensional head region.

The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, when the processor executes the program, the brain-computer-based computer program described in any of the above-mentioned processes is implemented by the processor. Interface for manipulator trajectory planning method.

The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements any of the above-mentioned brain-computer interface-based robotic arm trajectory planning methods.

The brain-computer interface-based robotic arm trajectory planning method, device, and electronic device provided by the present invention perform collision detection based on the three-dimensional head area and the first configuration list, and delete the robotic arm and the three-dimensional robotic arm in the first configuration list based on the collision detection result. There are collision 3D points in the head area and the joint angle data of the robot arm corresponding to the 3D points, so as to avoid the collision between the robot arm and the 3D head area during the process of implanting the brain-computer interface into the 3D head area, and improve the brain-computer interface. It can improve the implantation efficiency and safety of the robot, and avoid many production safety problems; by setting a more reasonable target access order based on the second configuration list, the access efficiency of the robot arm to multiple 3D points in the 3D head area can be improved, and then It can improve the implantation efficiency of the robot arm to implant the brain-computer interface into multiple 3D points in the 3D head area, and solve the problem that the movement of the robot arm in the highly cluttered 3D head area cannot be quickly and safely obtained in the existing technology. The simulated implantation trajectory formed, and then the technical problem of implanting the brain-computer interface into the three-dimensional head region based on the simulated implantation trajectory.

Description of drawings

In order to explain the present invention or the technical solutions in the prior art more clearly, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are the For some embodiments of the invention, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

1 is one of the schematic flow charts of the method for planning the trajectory of a robotic arm based on a brain-computer interface provided by the present invention;

2 is the second schematic flowchart of the method for planning the trajectory of a robotic arm based on a brain-computer interface provided by the present invention;

3 is a third schematic flowchart of a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention;

4 is a fourth schematic flowchart of a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention;

5 is a fifth schematic flowchart of a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention;

6 is a sixth schematic flowchart of a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention;

7 is a schematic structural diagram of a three-dimensional point cascade diagram in a specific embodiment of the present invention;

8 is a seventh schematic flowchart of a method for planning a trajectory of a robotic arm based on a brain-computer interface provided by the present invention;

9 is a schematic structural diagram of a robotic arm trajectory planning device of a brain-computer interface provided by the present invention;

FIG. 10 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention. , not all examples. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The following describes the method for planning a trajectory of a robotic arm based on a brain-computer interface of the present invention with reference to FIGS. 1 to 7 . As shown in FIG. 1 , the present invention provides a method for planning a trajectory of a robotic arm based on a brain-computer interface, including:

Step S1: Obtain a first configuration list, where the first configuration list includes each 3D point in the 3D head area and the manipulator arm joint angle data at each 3D point in the 3D head area.

Step S2: Perform collision detection based on the 3D head area and the first configuration list, and delete the 3D points where the robot collides with the 3D head area and the robot arm joint angle data corresponding to the 3D points in the first configuration list based on the collision detection result .

In step S3, the deleted first configuration list is used as the second configuration list, and the target access sequence of the robotic arm to all the three-dimensional points in the second configuration list is acquired.

Step S4, based on the second configuration list and the target access sequence, obtain a simulated implantation trajectory formed by the movement of the robotic arm in the 3D head area, and the simulated implantation trajectory is used to implant the brain-computer interface into the 3D head area.

The simulated implantation trajectory represents a motion trajectory formed by the simulated robotic arm moving in the three-dimensional head region based on the second configuration list and the target access sequence.

Further, based on the simulated implantation trajectory, the robotic arm is controlled to implant the brain-computer interface into each 3D point in the simulated implantation trajectory where the robotic arm does not collide with the three-dimensional head area, so as to realize the implantation of the brain-computer interface into the three-dimensional head. The specified head position in the head area.

From the above steps S1 to S4, the collision detection is performed based on the three-dimensional head area and the first configuration list, and based on the collision detection result, the three-dimensional points where the robotic arm collides with the three-dimensional head area and the corresponding three-dimensional points in the first configuration list are deleted. The joint angle data of the robotic arm can avoid the collision between the robotic arm and the three-dimensional head area during the process of implanting the brain-computer interface into the three-dimensional head area, improve the implantation efficiency and safety of the brain-computer interface, and avoid causing a lot of production. Security issue; by setting a more reasonable target access order based on the second configuration list, the access efficiency of the robotic arm to multiple 3D points in the 3D head area can be improved, and the implantation of the brain-computer interface into the 3D head area by the robotic arm can be improved. The implantation efficiency of multiple 3D points in the head area solves the problem that the simulated implantation trajectory formed by the movement of the robotic arm in the highly cluttered 3D head area cannot be quickly and safely obtained in the prior art. Technical issues of implanting a brain-computer interface into the three-dimensional head region.

In one embodiment, as shown in FIG. 2 , the above step S1 includes steps S11 to S13, wherein:

Step S11 , acquiring the robot arm pose data at each three-dimensional point in the three-dimensional head region of the robot arm, where the robot arm pose data includes the three-dimensional coordinate value of the robot arm and the rotation angle of the coordinate axis.

Wherein, the three-dimensional coordinate value includes x coordinate, y coordinate and z coordinate. The coordinate axis rotation angle includes an x-axis rotation angle, a y-axis rotation angle, and a z-axis rotation angle. The x-axis rotation angle represents the rotation angle around the x-axis. The y-axis rotation angle represents the rotation angle of the rotation around the y-axis. The z-axis rotation angle represents the rotation angle of the rotation around the z-axis.

Step S12 , based on the kinematic inverse solution operation strategy, the robot arm pose data is converted into the robot arm joint rotation angle data, and the robot arm joint rotation angle data includes a plurality of robot arm joints and the rotation angle of each robot arm joint.

Optionally, the manipulator joint rotation angle data includes six manipulator joints and the rotation angle of each manipulator joint.

It should be noted that the process of converting the pose data of the manipulator into the joint angle data of the manipulator can be called solving the inverse kinematics of the manipulator. The kinematic inverse solution operation strategy (that is, the solution method of the inverse kinematics of the manipulator) includes but is not limited to the analytical method, the geometric projection method, and the combination of the analytical method and the geometric projection method. Among them, the analytical method includes a method of converting the pose data of the manipulator into the data of the joint angle of the manipulator by using a pre-built kinematic inverse solution calculation model.

Step S13, obtaining a first configuration list based on each 3D point in the 3D head area and the manipulator joint rotation angle data corresponding to each 3D point.

In one embodiment, one of the three coordinate axis rotation angles is a variable whose value range is not limited. The coordinate axis rotation angle is assigned as a preset value set, wherein the preset value set is a set of values with a discretization step size.

It should be noted that, since one of the three coordinate axis rotation angles is a variable, and there is a variable in the input manipulator pose data, a definite manipulator joint angle data cannot be output. The variable coordinate axis rotation angle is assigned as a preset value set, which can not only ensure that the value of the coordinate axis rotation angle can change within a certain range, but also ensure that a certain robot arm is output based on the input robot arm pose data Joint angle data.

In one embodiment, since the value of the rotation angle of one of the coordinate axes is within the range of the preset value set, that is, it has multiple values, the pose data of the robotic arm corresponding to each three-dimensional point is multiple sets, which are determined by the mechanical The joint angle data of the manipulator obtained by the conversion of the arm pose data are also multiple sets. At this time, the multiple sets of manipulator joint angle data corresponding to each three-dimensional point can be expressed by the following formula (1):

表示多组机械臂关节转角数据，

表示第1组机械臂关节转角数据，

表示第2组机械臂关节转角数据，

表示第j组机械臂关节转角数据，

表示第m组机械臂 关节转角数据，m表示机械臂关节转角数据的个数。 in,

Represents multiple sets of joint angle data of the manipulator,

Indicates the joint angle data of the first group of manipulators,

Indicates the joint angle data of the second group of manipulators,

represents the joint angle data of the jth group of manipulators,

Indicates the mth group of joint angle data of the manipulator, and m represents the number of manipulator joint angle data.

In the case that the joint angle data of the manipulator includes six manipulator joints and the angle of each manipulator joint, each group of manipulator joint angle data can be represented by the following formula (2):

表示第j组机械臂关节转角数据，

表示第1个机械臂关节的转角，

表 示第2个机械臂关节的转角，

表示第3个机械臂关节的转角，

表示第4个机械臂关节的 转角，

表示第5个机械臂关节的转角，

表示第6个机械臂关节的转角。 in,

represents the joint angle data of the jth group of manipulators,

Indicates the rotation angle of the first robotic arm joint,

Indicates the rotation angle of the second robotic arm joint,

Indicates the rotation angle of the third robotic arm joint,

Indicates the rotation angle of the fourth robotic arm joint,

Indicates the rotation angle of the fifth robotic arm joint,

Indicates the rotation angle of the sixth robotic arm joint.

Preferably, the collision detection process in step S2 is implemented based on a 2-Opt algorithm, where 2-opt is an abbreviation for 2-optimization.

In one embodiment, as shown in FIG. 3 , the above step S2 includes steps S21 to S24, wherein:

Step S21 , acquiring a first bounding box corresponding to each three-dimensional point of the manipulator in the first configuration list and a second bounding box corresponding to the three-dimensional head region.

Among them, the bounding box is an algorithm for solving the optimal bounding space of discrete point sets. The objects in this embodiment are the robotic arm and the three-dimensional head area.

Step S22, for each first bounding box, determine whether the first bounding box intersects with the second bounding box.

Step S23 , when the first bounding box intersects with the second bounding box, determine whether the robotic arm intersects the three-dimensional head region at the three-dimensional point corresponding to the first bounding box.

Step S24 , in the case that the manipulator intersects the three-dimensional head region at the three-dimensional point corresponding to the first bounding box, delete the three-dimensional point and its corresponding manipulator joint angle data in the first configuration list.

In the above steps S21 to S24, by judging whether the first bounding box intersects with the second bounding box, and when the first bounding box intersects with the second bounding box, it is judged that the robotic arm is at the three-dimensional point corresponding to the first bounding box Whether it intersects with the 3D head area, because the intersection algorithm of the first bounding box and the second bounding box is simpler than the intersection algorithm of the manipulator object at the 3D point and the 3D head area object, this method can quickly rule out Disjoint objects, thereby simplifying the collision detection step and speeding up the collision detection efficiency.

In one embodiment, as shown in FIG. 4 , the above step S21 includes steps S211 to S213, wherein:

Step S211, obtaining the coordinate extreme value data of the manipulator at each three-dimensional point of the manipulator in the first configuration list, where the coordinate extreme value data of the manipulator includes a plurality of manipulator joints of the manipulator and the coordinate pole of each manipulator joint. value data.

The coordinate extreme value data includes the x coordinate extreme value, the y coordinate extreme value, and the z coordinate extreme value of the manipulator joint. The x-coordinate extrema include the largest x-coordinate and the smallest x-coordinate. The y-coordinate extrema include the largest y-coordinate and the smallest y-coordinate. The z-coordinate extrema include the maximum z-coordinate and the minimum z-coordinate.

Step S212 , obtaining a first bounding box corresponding to each manipulator joint in the manipulator at each three-dimensional point in the first configuration list based on the coordinate extreme value data of the manipulator.

Specifically, for each three-dimensional point in the first configuration list, in the case where the manipulator is located at the three-dimensional point, obtain the location where each manipulator joint is located based on the coordinate extreme value data of each manipulator joint in the manipulator. The first bounding box corresponding to the three-dimensional point.

Step S213: Obtain the regional coordinate extreme value data of the three-dimensional head region, and obtain the second bounding box corresponding to the three-dimensional head region based on the regional coordinate extreme value data.

The area coordinate extreme value data includes the x coordinate extreme value, the y coordinate extreme value and the z coordinate extreme value of the midpoint of the three-dimensional head area.

In one embodiment, the first bounding box and the second bounding box are axis-aligned bounding boxes (AABBs), which are the smallest hexahedron containing the object, with edges parallel to the axis. The first bounding box is the smallest hexahedron containing the joints of the manipulator. The second bounding box is the smallest hexahedron containing the three-dimensional head region.

Through the above embodiment, the first bounding box and the second bounding box are constructed by using the axis-aligned bounding box. Since the axis-aligned bounding box only needs six basic geometric elements of the object to complete the construction of the bounding box, on the one hand, this method is adopted. This method can reduce the calculation amount of the bounding box construction process and improve the calculation efficiency of the bounding box construction process. Computational efficiency of the box intersection operation.

In one embodiment, the above-mentioned steps S22 to S24 further include: for each three-dimensional point in the first configuration list, when the manipulator is located at the three-dimensional point, judging that each manipulator joint in the manipulator is located at the three-dimensional point. Whether the first bounding box corresponding to the three-dimensional point intersects the second bounding box, in the case that the first bounding box corresponding to the three-dimensional point intersects with the second bounding box of any one of the manipulator joints in the robotic arm, Determine whether the robotic arm intersects the three-dimensional head area at the three-dimensional point, and in the case where the robotic arm intersects the three-dimensional head area at the three-dimensional point, delete the three-dimensional point and its Corresponding manipulator joint angle data.

In one embodiment, as shown in FIG. 5 , the above step S3 includes steps S31 to S37, wherein:

Step S31, obtaining the current iteration count and the current access sequence of the robotic arm to all the three-dimensional points in the second configuration list.

Step S32, based on the current access sequence and the three-dimensional coordinates of each three-dimensional point in the second configuration list, obtain a current access path corresponding to the current access sequence.

Step S33: Acquire a local access sequence between any two non-adjacent three-dimensional points in the current access sequence, and perform inversion processing on the local access sequence in the current access sequence to obtain a first access sequence.

For example, the current access sequence is (A, B, C, D, E, F, G), select the local access sequence between any two non-adjacent 3D points B and 3D points E in the current access sequence (B, C, D, E), invert the local access sequence in the current access sequence, and the current access sequence after inversion processing is the first access sequence (A, E, D, C, B, F, G).

Step S34, based on the first access sequence shown and the three-dimensional coordinates of each three-dimensional point in the second configuration list, obtain a first access path corresponding to the first access sequence. The three-dimensional coordinates include x-coordinates, y-coordinates and z-coordinates.

Step S35, based on the current access path and the first access path, update the current access sequence and the current iteration count.

In one embodiment, the path lengths of the current access path and the first access path are obtained and compared, and in the case that the path length of the first access path is smaller than the path length of the current access path, the first access order is used as the updated current access path. access order, and assigns the current iteration count to zero. When the path length of the first access path is greater than or equal to the path length of the current access path, the current access sequence is not updated, and a value of 1 is added to the current iteration count to obtain the updated current iteration count.

Further, the Euclidean distance is used to calculate the path length of the current access path and the first access path, and the path length of the access path is shown in formula (3):

表示路径长度，

表示访问路径的起点三维坐标，且

。

表 示访问路径的终点三维坐标，且

。 in,

represents the path length,

represents the three-dimensional coordinates of the starting point of the access path, and

.

represents the three-dimensional coordinates of the end point of the access path, and

.

Step S36, the updated current access sequence is regarded as the new current access sequence, and the updated current iteration count is regarded as the new current iteration count.

In step S37, the above steps are repeatedly performed until the current iteration count reaches the maximum number of iterations, and the finally obtained current access sequence is used as the target access sequence.

From the above steps S31 to S37, the first access sequence is obtained by inverting the partial access sequence in the current access sequence through multiple iterations, and the path lengths of the current access path and the first access path are compared. When the length is less than the path length of the current access path, the first access order is used as the updated current access order, so that the current access path corresponding to the final current access order is the shortest, and the final current access order is obtained. As the target access sequence, the access efficiency of the robotic arm to multiple three-dimensional points in the three-dimensional head region can be improved, and the implantation of the brain-computer interface by the robotic arm to multiple three-dimensional points in the three-dimensional head region can be improved. efficiency.

Preferably, the optimization process of the second configuration list in step S4 is implemented based on the Dijkstra algorithm (ie, the Dijkstra algorithm), so as to obtain the target configuration list.

In one embodiment, as shown in FIG. 6 , the above step S4 includes steps S41 to S45, wherein:

Step S41, constructing a three-dimensional point cascade diagram based on the target access sequence and the second configuration list; the three-dimensional point cascade diagram includes a plurality of cascaded three-dimensional point layers and a plurality of vertices corresponding to each three-dimensional point layer, wherein each three-dimensional point cascade The point layer corresponds to a 3D point, and multiple vertices represent multiple sets of joint angle data of the manipulator corresponding to the 3D point.

It should be noted that the multiple vertices corresponding to each three-dimensional point layer represent multiple sets of joint rotation angle data of the manipulator corresponding to each three-dimensional point of the manipulator.

Step S42, traverse each three-dimensional point layer in the three-dimensional point cascade diagram, select a vertex in each of the two adjacent three-dimensional point layers, and obtain the position between the two adjacent three-dimensional point layers of the robotic arm based on the selected two vertexes. moving paths and moving costs.

In step S43, the previous step is repeated until all vertices in the three-dimensional point cascade graph are traversed, and all moving paths and moving costs are finally obtained.

Step S44 , traverse each 3D point layer in the 3D point cascade diagram, screen out target movement paths from multiple movement paths corresponding to two adjacent 3D point layers based on the movement cost, and obtain multiple target movement paths.

Preferably, the moving path with the smallest moving cost is selected from the multiple moving paths corresponding to the two adjacent three-dimensional point layers as the target moving path, thereby reducing the need for the robotic arm to move in the three-dimensional head region for implantation of the brain-computer interface. input cost.

Step S45 , obtaining a target configuration list based on the multiple target movement paths and the second configuration list, and obtaining a simulated implantation trajectory formed by the robotic arm in the three-dimensional head region based on the target configuration list.

Wherein, the target configuration list includes a plurality of three-dimensional points arranged in the order of access to the target and the data of the joint rotation angle of the manipulator corresponding to each three-dimensional point.

Further, the simulated implantation trajectory formed by the robotic arm in the three-dimensional head region can be expressed by the following formula (4):

表示模拟植入轨迹，

表示起始三维点，

表示终点三维 点，

表示起始三维点与终点三维点之间的其他三维点的速度数据，该速度数据包括其他 三维点的加速度和速度。K表示其他三维点的数量。 in,

represents the simulated implantation trajectory,

represents the starting 3D point,

represents the three-dimensional point of the end point,

Indicates the velocity data of other 3D points between the start 3D point and the end 3D point, and the velocity data includes the acceleration and velocity of the other 3D points. K represents the number of other three-dimensional points.

Preferably, an open-source robot motion planning algorithm and a target configuration list are used to generate a simulated implantation trajectory formed by the robotic arm in the three-dimensional head region. Among them, the open source robot motion planning algorithm is the OMPL algorithm (The Open MotionPlanning Library, OMPL).

In one embodiment, FIG. 7 is a schematic structural diagram of a three-dimensional point cascade diagram in a specific embodiment of the present invention. As shown in FIG. 7 , the three-dimensional point cascade diagram includes multiple cascaded three-dimensional point layers and each three-dimensional point layer. corresponding multiple vertices. The 3D point layer is composed of multiple vertices arranged vertically. The number of the 3D point layer is [1,n]. Each circle in Figure 7 represents a vertex, and the black solid circle on the left in Figure 7 represents the starting vertex. The black solid circle on the right represents the end vertex. The line between two adjacent vertices represents the movement path between the two adjacent three-dimensional point layers corresponding to the vertices. For example, the 3D point layer 1 and the 3D point layer 2 are two adjacent 3D point layers.

In one embodiment, as shown in FIG. 8 , the above step S42 includes steps S421 to S424, wherein:

Step S421 , acquiring the manipulator joint rotation angle data corresponding to the two selected vertices and the maximum rotation angle speed of the manipulator joint.

Step S422 , obtaining the manipulator joint angle difference based on the manipulator joint angle data corresponding to the two vertices, and obtaining the maximum joint angle time based on the manipulator joint angle difference and the maximum angle speed.

进行区分。例如，若机械臂关节的数量为六个，则

取值为[1,6]。 Wherein, the manipulator joint angle data includes a plurality of manipulator joints and a joint angle corresponding to each manipulator joint. Multiple robotic arm joints are numbered by joint

differentiate. For example, if the number of manipulator joints is six, then

The value is [1,6].

Step S423: Obtain the corner symbol switching cost of the robotic arm joint based on the robotic arm joint corner data corresponding to the two vertices; based on the corner symbol switching cost and a preset extra cost coefficient, obtain the robotic arm between two adjacent three-dimensional point layers. Additional cost of moving.

Step S424, based on the maximum joint rotation time and the extra cost, obtain the moving cost of the robotic arm moving between two adjacent three-dimensional point layers.

，基于机械臂关节

的关节 转角计算其对应的机械臂关节转角差值。基于机械臂关节

的最大转角角速度及其对应的 机械臂关节转角差值计算得到其对应的最大关节转角时间。基于机械臂关节

的关节转角 计算其对应的转角符号切换成本，并基于机械臂关节

对应的转角符号切换成本以及预设 额外成本系数，获取机械臂关节

在相邻两个三维点层中的两个顶点之间移动的额外成 本。基于机械臂关节

对应的最大关节转角时间和额外成本，计算得到机械臂关节

在相 邻两个三维点层中的两个顶点之间移动的移动成本。其中，机械臂关节

表示机械臂中任 意一个机械臂关节。 In one embodiment, for the robotic arm joints corresponding to the two vertices

, based on the robotic arm joint

The joint angle of the corresponding robot arm joint angle difference is calculated. Based on robotic arm joints

The maximum corner angular velocity of , and its corresponding joint corner difference of the manipulator are calculated to obtain its corresponding maximum joint corner time. Based on robotic arm joints

The joint angle of

Corresponding corner symbol switching cost and preset additional cost factor to obtain the joints of the robotic arm

The additional cost of moving between two vertices in two adjacent 3D point layers. Based on robotic arm joints

The corresponding maximum joint angle time and extra cost are calculated to obtain the robot arm joints

The movement cost of moving between two vertices in two adjacent 3D point layers. Among them, the mechanical arm joint

Indicates any robotic arm joint in the robotic arm.

Further, the calculation method of the moving cost of the robotic arm moving between two adjacent three-dimensional point layers is shown in the following formula (5):

表示机械臂关节的关节编号。

表示前一个三维点层中顶点对应的机械臂关节

的第一关节转角，

表示后一个三维 点层中顶点对应的机械臂关节

的第二关节转角。

表示相邻两个三维点层中顶 点对应机械臂关节

的机械臂关节转角差值。

表示机械臂关节

的最大转角角速度，

表示机械臂关节

的最大关节转角时间。

表示符号函数，

表示机械臂中机械臂关节

在相邻两个三维点层之间移动的转角符 号切换成本。 Among them, C represents the preset extra cost coefficient, which is a constant.

Indicates the joint number of the robot arm joint.

Indicates the robot arm joint corresponding to the vertex in the previous 3D point layer

The first joint angle of ,

Indicates the robot arm joint corresponding to the vertex in the latter 3D point layer

of the second joint angle.

Indicates that the vertices in the adjacent two 3D point layers correspond to the manipulator joints

The difference value of the manipulator joint rotation angle.

Represents a robotic arm joint

The maximum corner angular velocity of ,

Represents a robotic arm joint

The maximum joint rotation time.

represents a symbolic function,

Represents the robotic arm joints in the robotic arm

Corner symbol switching cost for moving between two adjacent 3D point layers.

的关节转角从负角度转到正角度或者从负角度转 到正角度的情况下，转角符号切换成本为1。在机械臂关节

的关节转角从负角度转到负角 度或者从正角度转到正角度的情况下，转角符号切换成本为0。即相邻两个三维点层中顶点 所对应的机械臂关节

具有不同的转角符号的情况下，机械臂从相邻两个三维点层之间移 动的移动成本更高。 It should be noted that in the robotic arm joint

When the joint angle of , changes from a negative angle to a positive angle or from a negative angle to a positive angle, the corner sign switching cost is 1. Joints in the robotic arm

When the joint angle of , changes from a negative angle to a negative angle or from a positive angle to a positive angle, the cost of switching the sign of the angle is 0. That is, the robot arm joints corresponding to the vertices in the adjacent two 3D point layers

With different sign of the corners, the moving cost of the manipulator from between two adjacent 3D point layers is higher.

A specific embodiment is provided below to further illustrate the brain-computer interface-based robotic arm trajectory planning method provided by the present invention, which specifically includes the following steps:

Step 1: Obtain the robot arm pose data at each 3D point in the 3D head area of the robot arm. The robot arm pose data includes the 3D coordinate value of the robot arm and the rotation angle of the coordinate axis; based on the kinematic inverse solution operation strategy Convert the robot arm pose data to the robot arm joint angle data. The robot arm joint angle data includes multiple robot arm joints and the rotation angle of each robot arm joint; based on each 3D point and each 3D point in the 3D head area Corresponding joint angle data of the manipulator is obtained to obtain the first configuration list.

Step 2: Obtain the first bounding box corresponding to each 3D point of the robotic arm in the first configuration list and the second bounding box corresponding to the 3D head area; for each first bounding box, determine whether the first bounding box is Intersect with the second bounding box; when the first bounding box intersects with the second bounding box, determine whether the robotic arm intersects the three-dimensional head area at the three-dimensional point corresponding to the first bounding box; when the robotic arm is in the first bounding box In the case where the 3D point corresponding to the box intersects with the 3D head area, the 3D point and its corresponding robot arm joint rotation angle data are deleted from the first configuration list. Use the deleted first configuration list as the second configuration list.

Step 3: Obtain the current iteration count and the current access order of the robotic arm to all the three-dimensional points in the second configuration list; based on the current access order and the three-dimensional coordinates of each three-dimensional point in the second configuration list, obtain the current access order corresponding to the current access order. Access path; obtain the local access sequence between any two non-adjacent three-dimensional points in the current access sequence, and reverse the local access sequence in the current access sequence to obtain the first access sequence; based on the first access sequence shown and the three-dimensional coordinates of each three-dimensional point in the second configuration list, obtain the first access path corresponding to the first access sequence; based on the current access path and the first access path, update the current access sequence and the current iteration count; The access order is taken as the new current access order, and the updated current iteration count is taken as the new current iteration count; the above steps are repeated until the current iteration count reaches the maximum number of iterations, and the final current access order is taken as the target access order.

Step 4: Build a 3D point cascade graph based on the target access order and the second configuration list; the 3D point cascade graph includes multiple cascaded 3D point layers and multiple vertices corresponding to each 3D point layer, wherein each 3D point cascade The point layer corresponds to a 3D point, and multiple vertices represent multiple sets of manipulator joint angle data corresponding to the 3D point; traverse each 3D point layer in the 3D point cascade diagram, and select a vertex in each of the two adjacent 3D point layers , and based on the selected two vertices, the moving path and moving cost of the robotic arm moving between two adjacent three-dimensional point layers are obtained; the previous step is repeated until all vertices in the three-dimensional point cascade graph are traversed, and the final result is obtained. All movement paths and movement costs; traverse each 3D point layer in the 3D point cascade diagram, screen out the target movement paths from the multiple movement paths corresponding to two adjacent 3D point layers based on the movement cost, and obtain multiple target movements path; obtaining a target configuration list based on the multiple target movement paths and the second configuration list, and obtaining a simulated implantation trajectory formed by the robotic arm in the three-dimensional head region based on the target configuration list.

Step 5: Obtain the manipulator joint angle data corresponding to the two selected vertices and the maximum angle angular velocity of the manipulator joint; obtain the manipulator joint angle difference based on the manipulator joint angle data corresponding to the two vertices, and based on the manipulator arm joint angle difference Obtain the maximum joint corner time based on the joint corner difference and the maximum corner angular velocity; obtain the corner symbol switching cost of the manipulator joint based on the manipulator joint corner data corresponding to the two vertices; obtain the mechanical arm joint based on the corner symbol switching cost and the preset extra cost factor. The additional cost of the arm moving between two adjacent 3D point layers; based on the maximum joint rotation time and the additional cost, obtain the movement cost of the robot arm moving between two adjacent 3D point layers.

The following describes the robot trajectory planning device based on the brain-computer interface provided by the present invention. The robot trajectory planning device based on the brain-computer interface described below and the robot trajectory planning method based on the brain-computer interface described above may refer to each other correspondingly. .

As shown in FIG. 9 , the present invention provides a robotic arm trajectory planning device based on a brain-computer interface. The brain-computer interface-based robotic arm trajectory planning device 100 includes: a configuration acquisition module 10 for acquiring a first configuration list, a first The configuration list includes each 3D point in the 3D head area and the robot arm joint rotation angle data at each 3D point in the 3D head area. The collision detection module 20 is configured to perform collision detection based on the 3D head area and the first configuration list, and delete the 3D points where the robotic arm collides with the 3D head area and the machinery corresponding to the 3D points in the first configuration list based on the collision detection result Arm joint rotation angle data. The sequence determination module 30 is configured to use the deleted first configuration list as the second configuration list, and obtain the target access sequence of the robotic arm to all the three-dimensional points in the second configuration list. The trajectory generation module 40 is configured to obtain a simulated implantation trajectory formed by the movement of the robotic arm in the three-dimensional head region based on the second configuration list and the target access sequence, and the simulated implantation trajectory is used to implant the brain-computer interface into the three-dimensional head area.

In one embodiment, the configuration obtaining module 10 includes: a pose data obtaining unit, configured to obtain the pose data of the robotic arm at each three-dimensional point in the three-dimensional head region of the robotic arm, where the pose data of the robotic arm includes the robotic arm The three-dimensional coordinate value and axis rotation angle of . The rotation angle data acquisition unit is used to convert the robot arm pose data into the robot arm joint rotation angle data based on the kinematic inverse solution operation strategy. The robot arm joint rotation angle data includes a plurality of robot arm joints and the rotation angle of each robot arm joint. The configuration list obtaining unit is configured to obtain the first configuration list based on each 3D point in the 3D head area and the data of the manipulator joint rotation angle corresponding to each 3D point.

In one embodiment, the collision detection module 20 includes: a bounding box obtaining unit, configured to obtain a first bounding box corresponding to each 3D point of the manipulator in the first configuration list and a second bounding box corresponding to the 3D head area box. The first intersection judging unit is configured to, for each first bounding box, judge whether the first bounding box intersects with the second bounding box. The second intersection determination unit is configured to determine whether the robotic arm intersects the three-dimensional head region at the three-dimensional point corresponding to the first bounding box when the first bounding box intersects with the second bounding box. The 3D point deletion unit is used to delete the 3D point and its corresponding robot arm joint angle data in the first configuration list when the robot arm intersects the 3D head area at the 3D point corresponding to the first bounding box.

In one embodiment, the bounding box obtaining unit includes: a coordinate extreme value obtaining subunit, configured to obtain coordinate extreme value data of the manipulator at each three-dimensional point of the manipulator in the first configuration list, coordinate extreme value data of the manipulator Including multiple manipulator joints of the manipulator and the coordinate extreme value data of each manipulator joint; the first bounding box acquisition subunit is used to obtain each manipulator joint in the manipulator based on the coordinate extreme value data of the manipulator. A first bounding box corresponding to each 3D point in the configuration list. The second bounding box obtaining subunit is used to obtain the regional coordinate extreme value data of the three-dimensional head region, and obtain the second bounding box corresponding to the three-dimensional head region based on the regional coordinate extreme value data.

In one embodiment, the sequence determination module 30 includes an iterative data acquisition unit, configured to acquire the current iteration count and the current access sequence of the robotic arm to all the three-dimensional points in the second configuration list. The first path obtaining unit is configured to obtain the current access path corresponding to the current access sequence based on the current access sequence and the three-dimensional coordinates of each three-dimensional point in the second configuration list. The local order inversion unit is used to obtain the local access order between any two non-adjacent three-dimensional points in the current access order, and perform reversal processing on the local access order in the current access order to obtain the first access order. The second path obtaining unit is configured to obtain a first access path corresponding to the first access sequence based on the first access sequence shown and the three-dimensional coordinates of each three-dimensional point in the second configuration list. An iterative data updating unit, configured to update the current access sequence and the current iteration count based on the current access path and the first access path. The iteration data determination unit is configured to use the updated current access sequence as the new current access sequence, and use the updated current iteration count as the new current iteration count. The target sequence determination unit is used to repeatedly execute the above steps until the current iteration count reaches the maximum number of iterations, and the final obtained current access sequence is used as the target access sequence.

In one embodiment, the trajectory generation module includes: a cascade graph construction unit, configured to construct a 3D point cascade graph based on the target access order and the second configuration list; the 3D point cascade graph includes a plurality of cascaded 3D point layers and Each 3D point layer corresponds to multiple vertices, wherein each 3D point layer corresponds to a 3D point, and the multiple vertices represent multiple sets of mechanical arm joint rotation angle data corresponding to the 3D point. The first mobile computing unit is used to traverse each 3D point layer in the 3D point cascade graph, select a vertex in each of the two adjacent 3D point layers, and obtain the position of the robot arm in the adjacent two vertices based on the selected two vertices. The movement path and movement cost of moving between three 3D point layers. The second movement calculation unit is used to repeatedly execute the previous step until all vertices in the three-dimensional point cascade graph are traversed, and finally all movement paths and movement costs are obtained. The movement path screening unit is used to traverse each 3D point layer in the 3D point cascade diagram, and based on the movement cost, select the target movement path from the multiple movement paths corresponding to the two adjacent 3D point layers, and obtain multiple target movement paths. path. The simulated trajectory generation unit is configured to obtain a target configuration list based on the multiple target moving paths and the second configuration list, and obtain a simulated implantation trajectory formed by the robotic arm in the three-dimensional head region based on the target configuration list.

In one embodiment, the first movement computing unit includes: a joint data acquisition subunit, configured to acquire the manipulator joint rotation angle data corresponding to the two selected vertices and the maximum rotation angle speed of the manipulator joint. The corner time acquisition subunit is used to obtain the difference of the joint corners of the robot arm based on the joint corner data of the robot arm corresponding to the two vertices, and to obtain the maximum joint corner time based on the difference of the joint corners of the robot arm and the maximum corner angular velocity. The additional cost acquisition subunit is used to obtain the corner sign switching cost of the manipulator joint based on the manipulator joint angle data corresponding to the two vertices; Additional cost of moving between 3D point layers. The movement cost acquisition sub-unit is used to obtain the movement cost of the robot arm moving between two adjacent 3D point layers based on the maximum joint rotation time and additional cost.

FIG. 10 illustrates a schematic diagram of the physical structure of an electronic device. As shown in FIG. 10 , the electronic device may include: a processor (processor) 1010, a communication interface (Communications Interface) 1020, a memory (memory) 1030, and a communication bus 1040, The processor 1010 , the communication interface 1020 , and the memory 1030 communicate with each other through the communication bus 1040 . The processor 1010 can call the logic instructions in the memory 1030 to execute a brain-computer interface-based robotic arm trajectory planning method. The method includes: acquiring a first configuration list, where the first configuration list includes each three-dimensional point in the three-dimensional head region and the joint angle data of the robot arm at each 3D point in the 3D head area; perform collision detection based on the 3D head area and the first configuration list, and delete the robot arm and the first configuration list based on the collision detection result. There are collision 3D points in the 3D head area and the joint angle data of the manipulator arm corresponding to the 3D points; take the deleted first configuration list as the second configuration list, and obtain the information about the robot arm for all the 3D points in the second configuration list. Target access sequence: Based on the second configuration list and the target access sequence, a simulated implantation trajectory formed by the movement of the robotic arm in the 3D head area is obtained, and the simulated implantation trajectory is used to implant the brain-computer interface into the 3D head area.

In addition, the above-mentioned logic instructions in the memory 1030 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

In another aspect, the present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored, and the computer program is implemented by a processor to execute the brain-computer interface-based robotic arm trajectory provided by the above methods when the computer program is executed. a planning method, the method comprising: acquiring a first configuration list, where the first configuration list includes each three-dimensional point in the three-dimensional head area and the manipulator joint rotation angle data of the manipulator at each three-dimensional point in the three-dimensional head area; Perform collision detection based on the 3D head area and the first configuration list, and delete the 3D points where the robot collides with the 3D head area and the joint angle data of the robot arm corresponding to the 3D points in the first configuration list based on the result of the collision detection; The processed first configuration list is used as the second configuration list, and the target access order of the robotic arm to all 3D points in the second configuration list is obtained; based on the second configuration list and the target access order, the robot arm in the 3D head area is obtained The simulated implantation trajectory formed by the middle movement is used to implant the brain-computer interface into the three-dimensional head region.

The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on this understanding, the above-mentioned technical solutions can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A mechanical arm track planning method based on a brain-computer interface is characterized by comprising the following steps:

acquiring a first configuration list, wherein the first configuration list comprises each three-dimensional point in a three-dimensional head area and mechanical arm joint corner data of a mechanical arm at each three-dimensional point in the three-dimensional head area;

performing collision detection based on the three-dimensional head area and the first configuration list, and deleting three-dimensional points where the mechanical arm collides with the three-dimensional head area and mechanical arm joint corner data corresponding to the three-dimensional points in the first configuration list based on a collision detection result;

taking the deleted first configuration list as a second configuration list, and acquiring a target access sequence of the mechanical arm to all three-dimensional points in the second configuration list;

and acquiring a simulated implantation track formed by the movement of the mechanical arm in the three-dimensional head region based on the second configuration list and the target access sequence, wherein the simulated implantation track is used for implanting a brain-computer interface into the three-dimensional head region.

2. The brain-computer interface based mechanical arm trajectory planning method according to claim 1, wherein the obtaining a first configuration list comprises:

acquiring mechanical arm position and pose data of the mechanical arm at each three-dimensional point in the three-dimensional head area, wherein the mechanical arm position and pose data comprise three-dimensional coordinate values and coordinate axis rotation angles of the mechanical arm;

converting the mechanical arm pose data into mechanical arm joint corner data based on a kinematics inverse solution operation strategy, wherein the mechanical arm joint corner data comprises a plurality of mechanical arm joints and corners of each mechanical arm joint;

and obtaining the first configuration list based on each three-dimensional point in the three-dimensional head area and the mechanical arm joint corner data corresponding to each three-dimensional point.

3. The brain-computer interface-based mechanical arm trajectory planning method according to claim 1 or 2, wherein the performing collision detection based on the three-dimensional head region and the first configuration list, and deleting three-dimensional points where a mechanical arm collides with the three-dimensional head region and mechanical arm joint rotation angle data corresponding to the three-dimensional points in the first configuration list based on a collision detection result comprises:

acquiring a first bounding box corresponding to each three-dimensional point in the first configuration list and a second bounding box corresponding to the three-dimensional head region of the mechanical arm;

for each of the first bounding boxes, determining whether the first bounding box intersects the second bounding box;

under the condition that the first enclosure box is intersected with the second enclosure box, judging whether a mechanical arm is intersected with the three-dimensional head area at a three-dimensional point corresponding to the first enclosure box;

and under the condition that the three-dimensional points corresponding to the first enclosure box of the mechanical arm are intersected with the three-dimensional head area, deleting the three-dimensional points and the corresponding mechanical arm joint rotation angle data in the first configuration list.

4. The brain-computer interface based mechanical arm trajectory planning method according to claim 3, wherein the obtaining a first bounding box corresponding to the mechanical arm at each three-dimensional point in the first configuration list and a second bounding box corresponding to the three-dimensional head region comprises:

acquiring mechanical arm coordinate extreme value data of the mechanical arm at each three-dimensional point in the first configuration list, wherein the mechanical arm coordinate extreme value data comprises a plurality of mechanical arm joints of the mechanical arm and coordinate extreme value data of each mechanical arm joint;

acquiring a first enclosure box corresponding to each mechanical arm joint in the mechanical arm at each three-dimensional point in the first configuration list based on the mechanical arm coordinate extreme value data;

and acquiring the region coordinate extreme value data of the three-dimensional head region, and acquiring a second enclosure box corresponding to the three-dimensional head region based on the region coordinate extreme value data.

5. The brain-computer interface based mechanical arm trajectory planning method according to claim 1 or 2, wherein the obtaining of the target access order of the mechanical arm to all three-dimensional points in the second configuration list comprises:

acquiring a current iteration count and a current access sequence of the mechanical arm to all three-dimensional points in the second configuration list;

acquiring a current access path corresponding to the current access sequence based on the current access sequence and the three-dimensional coordinates of each three-dimensional point in the second configuration list;

acquiring a local access sequence between any two non-adjacent three-dimensional points in the current access sequence, and turning over the local access sequence in the current access sequence to obtain a first access sequence;

acquiring a first access path corresponding to the first access sequence based on the first access sequence and the three-dimensional coordinates of each three-dimensional point in the second configuration list;

updating the current access order and the current iteration count based on the current access path and the first access path;

taking the updated current access sequence as a new current access sequence, and taking the updated current iteration count as a new current iteration count;

and repeatedly executing the steps until the current iteration count reaches the maximum iteration number, and taking the finally obtained current access sequence as the target access sequence.

6. The brain-computer interface-based mechanical arm trajectory planning method according to claim 1 or 2, wherein the obtaining of the simulated implantation trajectory formed by the movement of the mechanical arm in the three-dimensional head region based on the second configuration list and the target access sequence comprises:

constructing a three-dimensional point cascade graph based on the target access sequence and the second configuration list; the three-dimensional point cascade graph comprises a plurality of cascade three-dimensional point layers and a plurality of vertexes corresponding to each three-dimensional point layer, wherein each three-dimensional point layer corresponds to one three-dimensional point, and the vertexes represent a plurality of groups of mechanical arm joint corner data corresponding to the three-dimensional points;

traversing each three-dimensional point layer in the three-dimensional point cascade graph, respectively selecting a vertex in two adjacent three-dimensional point layers, and acquiring a moving path and a moving cost of the mechanical arm moving between the two adjacent three-dimensional point layers based on the two selected vertices;

repeatedly executing the previous step until all vertexes in the three-dimensional point cascade graph are traversed, and finally obtaining all moving paths and the moving cost;

traversing each three-dimensional point layer in the three-dimensional point cascade graph, and screening a target moving path from a plurality of moving paths corresponding to two adjacent three-dimensional point layers based on the moving cost to obtain a plurality of target moving paths;

and acquiring a target configuration list based on a plurality of target moving paths and the second configuration list, and acquiring a simulated implantation track formed by the mechanical arm in the three-dimensional head area based on the target configuration list.

7. The brain-computer interface-based mechanical arm trajectory planning method according to claim 6, wherein the obtaining of the movement path and the movement cost of the mechanical arm between the two adjacent three-dimensional point layers based on the two selected vertices comprises:

acquiring mechanical arm joint corner data corresponding to the two selected vertexes and the maximum corner angular speed of the mechanical arm joint;

acquiring a mechanical arm joint corner difference value based on mechanical arm joint corner data corresponding to the two vertexes, and acquiring maximum joint corner time based on the mechanical arm joint corner difference value and the maximum corner angular speed;

acquiring corner symbol switching cost of the mechanical arm joint based on the mechanical arm joint corner data corresponding to the two vertexes; acquiring the extra cost of the mechanical arm moving between the two adjacent three-dimensional point layers based on the corner symbol switching cost and a preset extra cost coefficient;

and acquiring the movement cost of the mechanical arm moving between the two adjacent three-dimensional point layers based on the maximum joint corner time and the additional cost.

8. A mechanical arm track planning device based on a brain-computer interface is characterized by comprising:

the configuration acquisition module is used for acquiring a first configuration list, wherein the first configuration list comprises each three-dimensional point in a three-dimensional head area and mechanical arm joint corner data of a mechanical arm at each three-dimensional point in the three-dimensional head area;

the collision detection module is used for performing collision detection based on the three-dimensional head area and the first configuration list, and deleting three-dimensional points where the mechanical arm collides with the three-dimensional head area and mechanical arm joint corner data corresponding to the three-dimensional points in the first configuration list based on a collision detection result;

the sequence determining module is used for taking the deleted first configuration list as a second configuration list and acquiring the target access sequence of the mechanical arm to all three-dimensional points in the second configuration list;

and the track generation module is used for acquiring a simulated implantation track formed by the movement of the mechanical arm in the three-dimensional head area based on the second configuration list and the target access sequence, wherein the simulated implantation track is used for implanting a brain-computer interface into the three-dimensional head area.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the brain-computer interface based robotic arm trajectory planning method of any one of claims 1 to 7.

10. A non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the brain-computer interface based mechanical arm trajectory planning method according to any one of claims 1 to 7.

Images