Autonomous Obstacle Avoidance Method for Mobile Robotic Arms Based on Zero-Space Control and Intention Guidance

Abstract

This invention discloses an autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance. It constructs a multi-layered local environmental perception point cloud map using point cloud data, generating a multi-layered perception range from the inside out, and reducing the sensitivity of the outer perception range. By determining the urgency of obstacle avoidance, the multi-layered local environmental perception, and the desensitized area point cloud map, a lateral obstacle avoidance linear velocity of the mobile chassis is generated. Combined with the maximum permissible lateral movement speed of the mobile chassis, the final lateral obstacle avoidance speed accepted by the mobile chassis is obtained. The velocity direction of the end effector projected onto the ground is defined using the mobile robotic arm's end effector, and a compensation gain coefficient and chassis deflection angle are used to generate a collaborator's motion intent following compensation angular velocity. The collaborator's motion intent following speed is obtained using the compensation angular velocity and the lateral obstacle avoidance speed. A kinematic model of the mobile robotic arm is constructed, and combined with the collaborator's motion intent following speed, autonomous obstacle avoidance of the mobile chassis in human-machine collaboration is achieved.

Description

Technical Field

[0001] This invention belongs to the field of robot motion control technology, specifically relating to an autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance. Background Technology

[0002] The integration of perception, manipulation, and mobility capabilities has become a research hotspot and development direction in robotics technology in recent years. Mobile robotic arms, composed of a manipulator and a mobile platform, possess both flexible movement and dexterous manipulation capabilities, and are increasingly being applied in industries, agriculture, and services, showing broad application prospects. However, the complex dynamic factors of the nonlinear strong coupling between the mobile platform and the manipulator directly affect the motion control performance of the mobile robotic arm. Developing efficient and advanced autonomous motion controllers remains a major challenge in realizing its application potential. In human-robot collaborative scenarios, considering the differences in motion characteristics between the mobile platform and the manipulator, and coordinating redundant degrees of freedom to achieve autonomous obstacle avoidance and real-time motion control in unstructured environments, are key challenges facing human-robot collaborative operations.

[0003] Traditional methods for coordinated motion control of mobile robotic arms often focus on decoupling control strategies. This involves using a higher-level planner to directly assign task objectives to the mobile platform and the manipulator, then designing independent controllers for each to complete the operation. It's evident that this higher-level planner cannot achieve coordinated allocation of redundant degrees of freedom across the entire mobile robotic arm for multi-task synchronization; it can only be used for discrete sequential operation control. Another research direction in coordinated motion control of mobile robotic arms focuses on joint control strategies. This involves simultaneously completing primary and secondary tasks by solving the redundant kinematics of the mobile robotic arm, meeting the vehicle-arm coordinated motion control requirements for continuous operation. Currently, methods such as multi-task hierarchical control and weighted least squares have been applied to the coordinated control of redundant kinematics in mobile robotic arms. These methods use gradient descent to optimize specific indices, but gradient optimization methods struggle to control the degree of optimization and quantify the optimization indices, and they do not consider the autonomous obstacle avoidance problem of the mobile chassis. In terms of obstacle avoidance for mobile chassis, existing research is limited to the global and local obstacle avoidance algorithms of the mobile chassis itself, lacking the utilization of the overall motion information of the mobile robotic arm. This results in a large computational cost for its obstacle avoidance algorithm, poor obstacle avoidance effect, and a lack of consideration for the interaction experience of collaborators. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies and achieve autonomous obstacle avoidance of a mobile robotic arm with zero-space control and human intention guidance, this invention adopts the following technical solution:

[0005] An autonomous obstacle avoidance method for mobile robotic arms based on zero-space control and intent guidance includes the following steps:

[0006] Step 1: Acquire point cloud data and construct a multi-layered local environment perception point cloud map. Based on the multi-layered perception range of the point cloud map from the inside out, reduce the sensitivity of the outer layer perception range to meet the actual obstacle avoidance needs in human-machine collaboration scenarios.

[0007] Step 2: Generate lateral obstacle avoidance speed commands and collaborator motion intention following speed commands to meet the obstacle avoidance requirements of the mobile chassis in unstructured scenarios. The lateral obstacle avoidance speed command is generated by defining the obstacle avoidance weights of each layer of perception areas to determine the urgency of obstacle avoidance from the inside out. The lateral obstacle avoidance linear velocity of the mobile chassis is generated based on the multi-layer local environment perception and desensitized area point cloud map. Combined with the maximum allowable lateral movement speed of the mobile chassis, the final lateral obstacle avoidance speed accepted by the mobile chassis is obtained. The collaborator motion intention following speed command is generated by defining the velocity direction of the end effector projected on the ground by the velocity component of the mobile robotic arm end effector in its base coordinate system. The velocity direction is combined with the compensation gain coefficient and chassis deflection angle to generate the collaborator motion intention following compensation angular velocity. The obstacle avoidance speed required by the mobile chassis, i.e., the collaborator motion intention following speed, is obtained by using the compensation angular velocity and the lateral obstacle avoidance speed.

[0008] Step 3: Construct a kinematic model of the mobile robotic arm, and design an augmented Jacobian zero-space control law based on the collaborator's motion intention to follow the speed, so as to achieve autonomous obstacle avoidance of the mobile chassis under the premise of satisfying the main task of human-machine collaboration.

[0009] Furthermore, in step 1, a multi-level local environment perception model is established. Within the maximum detection range of the mobile robotic arm's lidar, the model's perception range forms multiple progressively increasing perception areas outward from the center point of the lidar, in order to determine the urgency of obstacle avoidance.

[0010] Furthermore, in step 1, considering the actual obstacle avoidance requirements in human-machine collaboration scenarios, when the human-machine system approaches a wall, due to the limited perception method of single-line lidar, long and continuous walls will reflect a large number of laser beams, causing the mobile chassis to regard the wall that it can approach as a huge obstacle and thus be unable to get close. To address this issue, the sensitivity reduction region is designed using the hierarchical characteristics of the multi-layer local perception model. With the center of the mobile chassis of the mobile robotic arm as the origin of the coordinate system, multiple sensitivity reduction regions are established. Each sensitivity reduction region contains at least one outer perception region, thereby ensuring that the perception of obstacles in the first perception ring with the highest obstacle avoidance urgency is not affected.

[0011] Furthermore, in step 1, by collecting a one-dimensional array containing angle and distance information of the target, a relationship between the array index and the azimuth angle of the sampling point is established, and the distance of the sampling point with a return distance of infinity is set as the maximum detection distance.

[0012] Furthermore, a circular area is generated with the installation center point of the mobile robotic arm. A tangent line is drawn from the center point of the LiDAR on the mobile robotic arm to this circle to generate a shielded area. The LiDAR recognizes that there are no obstacles in this area, thereby removing the installation area of ​​the robotic arm and the human-guided work area from the LiDAR's perception area.

[0013] Furthermore, since this obstacle avoidance method is applicable to human-machine collaborative operation scenarios, where collaborators provide collaborative guidance in the longitudinal (X-axis direction) of the omnidirectional mobile chassis, and the applicable mobile chassis is an omnidirectional mobile chassis with Mecanum wheels, which has the ability to translate in any direction, the local environmental perception information provided by the lidar only needs to provide obstacle avoidance guidance in the lateral (Y-axis direction) of the chassis. That is, when storing lidar data, only the distance data on the Y-axis in the lidar coordinate system needs to be stored.

[0014] Furthermore, in step 2, a lateral obstacle avoidance speed command is generated by setting obstacle avoidance weight values ​​α, β, and γ from the inside out, and:

[0015] α+β+γ=1

[0016] 0≤α≤β≤γ≤1

[0017] The lateral obstacle avoidance linear velocity of the mobile chassis is as follows:

[0018]

[0019]

[0020] Where Laserscan1*[i], Laserscan2*[i], and Laserscan3*[i] represent the obstacle data storage arrays of the three-layer sensing ring, respectively, and G r The desensitization coefficient is denoted by m1 and m2, which represent the number of elements in the first and second rectangular desensitization regions DPC1[] and DPC2[] from the outside in, respectively. obs This represents the lateral obstacle avoidance linear velocity of the mobile chassis generated based on multi-layered local environmental perception and the rectangular desensitized area point cloud map, which is a weighted composite of the obstacle avoidance velocities generated by each perception ring and the obstacle avoidance sensitivity compensation velocities generated by the rectangular desensitized area; v max Indicates the maximum permissible lateral movement speed of the mobile chassis; v y The final lateral obstacle avoidance speed accepted by the mobile chassis.

[0021] Furthermore, in step 2, a motion intention-following speed command for the collaborator is generated by defining the velocity of the mobile robotic arm end effector in its base coordinate system as v. ee Take its velocity components on the X and Y axes. Define the velocity direction of the end effector projected onto the ground:

[0022]

[0023] The collaborator's motion intention follows the compensated angular velocity ω f The definition is as follows:

[0024] ω f =k f ·(θ ee -θ ω )

[0025] Where, k f θ represents the compensation gain coefficient. ω Indicates the chassis offset angle;

[0026] To achieve autonomous obstacle avoidance by the mobile chassis during human-machine collaborative handling, the required obstacle avoidance speed for the mobile chassis is as follows:

[0027] v m =[0,v y ,ω f ] T .

[0028] Furthermore, in step 3, the kinematic equations of the Mecanum wheeled mobile robotic arm are constructed as follows:

[0029]

[0030] Among them, v e Let q represent the velocity vector of the mobile robotic arm's end effector in the task space, q = [q m T ,q A T ] T Represents the generalized spatial coordinates of the mobile robotic arm. and q A =[q1,q2,…,q n ] T Let x represent the joint space coordinates of the mobile platform and the robotic arm, respectively. m y m Represents the horizontal and vertical coordinates of the mobile platform. The rotation angles of the mobile platform are represented by q1, q2, ..., q. n This represents the angles of each joint of the robotic arm, where n represents the number of joints in the robotic arm. Let J(q,φ) represent the joint space velocity of the mobile robotic arm, and let J(q,φ) represent the Jacobian matrix of the mobile robotic arm, with its elements satisfying a linear relationship with φ.

[0031] Furthermore, in step 3, the augmented Jacobian zero-space control law is designed as follows:

[0032]

[0033] in, This represents the final output speed of the moving robotic arm joints. and These represent the first-priority collaborative task and the second-priority obstacle avoidance task of the mobile chassis, i.e., the primary and secondary tasks, respectively. This is to ensure the convergence of trajectory tracking errors of the mobile robotic arm's end effector and the mobile platform. and This refers to the speed commands issued to the end effector of the mobile robotic arm and the mobile platform. The relationship between the above tasks and joint speeds can be expressed as follows:

[0034]

[0035] Where J1 = J(q,φ), J2 represents the Jacobian matrix of the secondary task. Since the constraints of the secondary task only apply to the moving chassis, its attitude is expressed as... The Jacobian matrix J2 for obstacle avoidance on the mobile chassis is as follows:

[0036] J2=(I 3×3 0 3×6 )

[0037] Among them, I 3×3 Represents a 3×3 diagonal matrix, 0 3×6 Represents a 3×6 matrix of all zeros;

[0038] In human-machine collaboration scenarios By integrating the mobile chassis with the additional obstacle avoidance speed v m and the moving chassis speed in the joint speed of collaborative tasks The formula is as follows:

[0039]

[0040] The advantages and beneficial effects of this invention are as follows:

[0041] The autonomous obstacle avoidance method for mobile robotic arms based on zero-space control and intent guidance of the present invention takes into account the actual obstacle avoidance situation in human-machine collaboration scenarios, can meet the obstacle avoidance requirements of mobile chassis in human-machine collaboration scenarios, can remove the installation area of ​​the robotic arm and the human-guided working area from the lidar perception area, can reduce the amount of lidar data storage, and ultimately realize autonomous obstacle avoidance of mobile chassis in human-machine collaborative handling process. Attached Figure Description

[0042] Figure 1 This is a flowchart of a method according to an embodiment of the present invention.

[0043] Figure 2aThis is a graph showing the end-point trajectory tracking error in an embodiment of the present invention.

[0044] Figure 2b This is a graph showing the trajectory tracking error of the mobile chassis in an embodiment of the present invention.

[0045] Figure 3 This is a line graph of the joint motion angle in an embodiment of the present invention.

[0046] Figure 4 This is a diagram showing the motion trajectory of a human-mobile robotic arm in an embodiment of the present invention. Detailed Implementation

[0047] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0048] like Figure 1 As shown, the autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance of the present invention includes the following steps:

[0049] Step 1: Preprocess single-line LiDAR point cloud data and construct a multi-level local environment perception point cloud map. Design an outer perception ring sensitivity reduction algorithm to meet the actual obstacle avoidance requirements in human-machine collaboration scenarios.

[0050] Step 1.1: The multi-layered local environment perception point cloud map is constructed as follows:

[0051] The data returned by a single-line lidar is a one-dimensional array containing angle and distance information, with the angle resolution set to R. θ =360 / n, the number of sampling points n = 360, and the angle θ in the lidar coordinate system corresponding to the array index value i = i·R θ After establishing the relationship between the array index and the azimuth angle of the sampling points, the data returned by the lidar is preprocessed. When storing detection data, the lidar will exclude data exceeding the maximum detection range. max The sampling point return distance is set to Inf (infinity). To avoid the infinity value affecting subsequent algorithm processing, the sampling point distance value with a return value of Inf is set to the maximum detection distance dis. max .

[0052] After completing the lidar data preprocessing, a multi-level local environment perception model is established. The model's perception area is divided into three levels. Within the lidar's maximum detection range, its detection area is divided into three circular regions with progressively increasing radii (R1 < R2 < R3). These form three perception rings with decreasing grayscale from the lidar center outwards. Higher grayscale indicates greater sensitivity to obstacles within the region and a higher degree of obstacle avoidance urgency. Red dots represent the distance the lidar laser beam returns after being blocked by an obstacle (an object composed of closed black solid lines) within that region. The detailed construction process of the multi-level local environment perception model is as follows: First, the lidar's perception area is redefined. Since it operates in a human-machine collaborative scenario, the robotic arm's installation area and the human-guided work area need to be removed from the lidar's perception area. Specifically, a ring with radius R is generated around the robotic arm's installation center point. arm A circle is drawn from the center point of the lidar, and a tangent line is drawn to this circle, generating an angle of... Within the shielded area, the distance values ​​detected by the lidar in this area all return the maximum detection range. max This indicates that there are no obstacles in that direction.

[0053] After redefining the sensing area, the angle and distance data (Laserscan[]) returned by the lidar in polar coordinates are converted to coordinates in Cartesian coordinates to obtain the X-axis coordinate matrix D. x ∈R 1×360 and the Y-axis coordinate matrix D y ∈R 1×360 Then, the obstacle data storage arrays corresponding to the three perception rings are defined as Laserscan1[], Laserscan2[], and Laserscan3[], respectively. It should be noted that since this obstacle avoidance method is applicable to human-machine collaborative operation scenarios, where collaborators provide collaborative guidance in the longitudinal (X-axis direction) of the omnidirectional mobile chassis, and the applicable mobile chassis is an omnidirectional mobile chassis with Mecanum wheels, which has the ability to translate in any direction, the local environmental perception information provided by the lidar only needs to provide obstacle avoidance guidance in the lateral (Y-axis direction) of the chassis. That is, when storing lidar data, only the distance data on the Y-axis in the lidar coordinate system needs to be stored. In summary, the pseudocode of the point cloud data storage algorithm for each perception ring is as follows.

[0054]

[0055] After the point cloud data of each sensing ring is stored, it is normalized to avoid its dimensions affecting subsequent processing. The normalization method is shown in Equation (1):

[0056]

[0057] Step 1.2: Based on Step 1.1, the sensitivity reduction algorithm for the outer sensing ring is designed as follows:

[0058] Considering the practical obstacle avoidance requirements in human-machine collaborative scenarios, when the human-machine system approaches a wall, due to the limitations of single-line LiDAR perception, long and continuous walls will reflect a large number of laser beams, causing the mobile chassis to perceive walls that it could approach as huge obstacles, thus preventing it from getting close. To address this issue, a sensitivity reduction region is designed using the hierarchical characteristics of a multi-layered local perception model. Using the center of the mobile chassis as the origin of the coordinate system, two rectangular regions with lengths l1 and l2 and widths h1 and h2 are established. These two rectangular regions only contain the second and third-layer perception loops, thus ensuring that obstacle perception in the first-layer perception loop, which has the highest obstacle avoidance urgency, remains unaffected. The pseudocode for the outer perception loop sensitivity reduction algorithm is as follows:

[0059]

[0060] Step 2: Design lateral obstacle avoidance speed commands and collaborator movement intention following speed commands to meet the obstacle avoidance requirements of mobile chassis in unstructured scenarios;

[0061] Step 2.1: Design the lateral obstacle avoidance speed command as follows;

[0062] First, the obstacle avoidance weight values ​​of each perception ring are defined as α, β and γ respectively. The relationship that the weight values ​​should satisfy is shown in Equation (2), which represents the increasing urgency of obstacle avoidance from the inside to the outside.

[0063]

[0064] The lateral obstacle avoidance linear velocity of the designed mobile chassis is shown in equation (3), where G r m1 and m2 are the desensitization coefficients, respectively, and the number of elements in the first and second layer rectangular desensitization regions DPC1[] and DPC2[], respectively. obs The lateral obstacle avoidance linear velocity of the mobile chassis, generated based on multi-layered local environmental perception and a rectangular desensitized area point cloud map, is a weighted composite of the obstacle avoidance velocity generated by each perception loop and the obstacle avoidance sensitivity compensation velocity generated by the rectangular desensitized area; v max v is the maximum permissible lateral movement speed of the mobile chassis. y The final lateral speed command received by the mobile chassis.

[0065]

[0066] Step 2.2: Design the following instructions for the collaborator's movement intention to follow the speed:

[0067] Define the velocity of the end effector of the mobile robotic arm in its base coordinate system as v. eeTake its velocity components on the X and Y axes. The velocity direction of the end effector projected onto the ground is defined as shown in equation (4):

[0068]

[0069] The collaborator's motion intention follows the compensated angular velocity ω f The definition is shown in equation (5), where k f To compensate for the gain coefficient, θ ω This refers to the chassis offset angle.

[0070] ω f =k f ·(θ ee -θ ω (5)

[0071] In summary, to achieve autonomous obstacle avoidance of the mobile chassis during human-machine collaborative handling, the obstacle avoidance speed required by the mobile chassis is shown in equation (6):

[0072] v m =[0, v y , ω f ] T (6)

[0073] Step 3: Construct a kinematic model of the mobile robotic arm and design an augmented Jacobian zero-space control law to achieve autonomous obstacle avoidance of the mobile chassis while satisfying the main task of human-machine collaboration.

[0074] Step 3.1: The kinematic equations of the Mecanum wheeled mobile robotic arm are constructed as follows:

[0075]

[0076] Among them, v e Let q represent the velocity vector of the mobile robotic arm's end effector in the task space, q = [q m T ,q A T ] T Represents the generalized spatial coordinates of the mobile robotic arm. and q A =[q1,q2,…,q n ] T Let x represent the joint space coordinates of the mobile platform and the robotic arm, respectively. m y m Represents the horizontal and vertical coordinates of the mobile platform. Let q represent the rotation angle of the mobile platform, q represent the joint angle of the robotic arm, and n represent the number of joints in the robotic arm. In this patent, n = 6. Let J(q,φ) represent the joint space velocity of the mobile robotic arm, and let J(q,φ) represent the Jacobian matrix of the mobile robotic arm, with its elements satisfying a linear relationship with φ.

[0077] Step 3.2: Design the augmented Jacobian zero-space control law as follows:

[0078]

[0079] In formula (8), The final output is the joint speed of the mobile robotic arm. and

[0080] These represent the primary and secondary tasks, respectively, which are the first-priority collaborative task and the second-priority obstacle avoidance task of the mobile chassis. To ensure the convergence of trajectory tracking errors at the end effector of the mobile robotic arm and the mobile platform, and The speed commands issued to the end effector of the mobile robotic arm and the mobile platform, and the relationship between the above tasks and joint speeds, can be expressed as shown in equation (9):

[0081]

[0082] Where J1 = J(q,φ), and J2 is the Jacobian matrix of the secondary task. Since the constraints of the secondary task only apply to the moving chassis, its attitude can be expressed as... Therefore, the Jacobian matrix J2 for obstacle avoidance of the mobile chassis is designed as shown in equation (10):

[0083] J2=(I 3×3 0 3×6 (10)

[0084] In the human-computer collaboration scenarios to which this invention is applicable By integrating the obstacle avoidance speed v of the mobile chassis in formula (6) m and the moving chassis speed in the joint speed of collaborative tasks The result is shown in equation (11):

[0085]

[0086] Example

[0087] The process of this invention, which uses a mobile robotic arm with zero-space control and human intention guidance for autonomous obstacle avoidance, is as follows: Figure 1 As shown, the specific implementation object is a mobile robotic arm consisting of a Mecanum wheeled mobile platform and a six-degree-of-freedom robotic arm, whose kinematic parameters are d. b=0.140m, d1=0.140m, a2=0.375m, a3=0.345m, d4=0.122m, d5=0.122m, d6=0.083m.

[0088] In the implementation of this invention, the desired trajectory of the collaborative task at the end effector of the mobile robotic arm is designed as follows: 0≤t≤20; the desired trajectory of the mobile chassis is designed as follows: 0≤t≤20. Based on the above settings, the implementation effect of the augmented Jacobian zero-space control law for the mobile robotic arm can be obtained. Figure 2a , Figure 2b This demonstrates that the augmented Jacobian zero-space control law of the mobile robotic arm can accurately track the primary and secondary task trajectories; Figure 3 This indicates that the joint movements during the exercise were relatively smooth; Figure 4 This demonstrates that the autonomous obstacle avoidance method for mobile robotic arms based on zero-space control and human intention guidance enables mobile robotic arms to effectively achieve obstacle avoidance on the mobile chassis in unstructured environments.

[0089] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance, characterized in that... Includes the following steps: Step 1: Acquire point cloud data and construct a multi-layered local environment perception point cloud map. Based on the multi-layered perception range of the point cloud map from the inside out, reduce the sensitivity of the outer layer perception range. Step 2: Generate lateral obstacle avoidance speed commands and collaborator motion intention following speed commands; The lateral obstacle avoidance speed commands are generated by defining the obstacle avoidance weights of each layer of perception areas to determine the urgency of obstacle avoidance from the inside out. Based on the multi-layer local environment perception and desensitized area point cloud map, the lateral obstacle avoidance linear velocity of the mobile chassis is generated. Combined with the maximum allowable lateral movement speed of the mobile chassis, the final lateral obstacle avoidance speed accepted by the mobile chassis is obtained. The collaborator motion intention following speed commands are generated by defining the velocity direction of the end effector projected on the ground by the velocity component of the end effector of the mobile robotic arm in its base coordinate system. The velocity direction is combined with the compensation gain coefficient and the chassis deflection angle to generate the collaborator motion intention following compensation angular velocity. Through the compensation angular velocity and the lateral obstacle avoidance speed, the obstacle avoidance speed required by the mobile chassis, i.e., the collaborator motion intention following speed, is obtained. Step 3: Construct a kinematic model of the mobile robotic arm, and design an augmented Jacobian zero-space control law based on the collaborator's motion intention to follow the speed, so as to achieve autonomous obstacle avoidance of the mobile chassis under the premise of satisfying the main task of human-machine collaboration.

2. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 1, characterized in that: In step 1, a multi-level local environment perception model is established. Within the maximum detection range of the lidar of the mobile robotic arm, the model's perception range forms multiple progressively increasing perception areas outward from the center point of the lidar.

3. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 2, characterized in that: In step 1, the sensitivity reduction region is designed using the hierarchical characteristics of the multi-level local perception model. The center of the mobile chassis of the mobile robotic arm is used as the origin of the coordinate system to establish multiple sensitivity reduction regions, each of which contains at least one outer perception region.

4. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 1, characterized in that: In step 1, by collecting a one-dimensional array containing angle and distance information of the target, the relationship between the array index and the azimuth angle of the sampling point is established, and the distance of the sampling point with a return distance of infinity is set as the maximum detection distance.

5. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 1, characterized in that: A circular area is generated from the installation center point of the mobile robotic arm. A tangent line is drawn from the center point of the LiDAR on the mobile robotic arm to this circle to generate a shielded area. The LiDAR recognizes that there are no obstacles in this area.

6. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 1, characterized in that: In human-machine collaborative operation scenarios, collaborators guide each other in the longitudinal direction of the omnidirectional mobile chassis. The applicable mobile chassis is an omnidirectional mobile chassis with Mecanum wheels, which has the ability to translate in any direction. Therefore, the local environmental perception information provided by the lidar only needs to provide obstacle avoidance guidance in the lateral direction of the chassis.

7. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 1, characterized in that: In step 2, a lateral obstacle avoidance speed command is generated by setting obstacle avoidance weight values ​​α, β, and γ from the inside out, and: α+β+γ=1 0≤α≤β≤γ≤1 The lateral obstacle avoidance linear velocity of the mobile chassis is as follows: Where Laserscan1*[i], Laserscan2*[i], and Laserscan3*[i] represent the obstacle data storage arrays of the three-layer sensing ring, respectively, and G r The desensitization coefficient is denoted by m1 and m2, which represent the number of elements in the first and second desensitization regions DPC1[] and DPC2[] from the outside in, respectively. obs This represents the lateral obstacle avoidance linear velocity of the mobile chassis generated based on multi-layered local environmental perception and desensitized area point cloud maps, which is a weighted composite of the obstacle avoidance velocities generated by each perception ring and the obstacle avoidance sensitivity compensation velocities generated by the rectangular desensitized area; v max Indicates the maximum permissible lateral movement speed of the mobile chassis; v y The final lateral obstacle avoidance speed accepted by the mobile chassis.

8. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 7, characterized in that: In step 2, a motion intention follow-speed command for the collaborator is generated by defining the velocity of the mobile robotic arm's end effector in its base coordinate system as v. ee Take its velocity components on the X and Y axes. Define the velocity direction of the end effector projected onto the ground: The collaborator's motion intention follows the compensated angular velocity ω f The definition is as follows: oh f =k f ·(θ ee -θ ω ) Where, k f θ represents the compensation gain coefficient. ω Indicates the chassis offset angle; The obstacle avoidance speeds required for the mobile chassis are as follows: v m =[0,v y ,ω f ] T 。 9. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 1, characterized in that: In step 3, the kinematic equations of the Mecanum wheeled mobile robotic arm are constructed as follows: Among them, v e Let q represent the velocity vector of the mobile robotic arm's end effector in the task space, q = [q m T ,q A T ] T Represents the generalized spatial coordinates of the mobile robotic arm. and q A =[q1,q2,…,q n ] T Let x represent the joint space coordinates of the mobile platform and the robotic arm, respectively. m y m Represents the horizontal and vertical coordinates of the mobile platform. The rotation angles of the mobile platform are represented by q1, q2, ..., q. n This represents the angles of each joint of the robotic arm, where n represents the number of joints in the robotic arm. Let J(q,φ) represent the joint space velocity of the mobile robotic arm, and let J(q,φ) represent the Jacobian matrix of the mobile robotic arm, with its elements satisfying a linear relationship with φ.

10. The autonomous obstacle avoidance method for a mobile robotic arm based on zero-space control and intent guidance according to claim 9, characterized in that: In step 3, the augmented Jacobian zero-space control law is designed as follows: in, This represents the final output speed of the moving robotic arm joints. and These represent the first-priority collaborative task and the second-priority obstacle avoidance task of the mobile chassis, i.e., primary and secondary tasks. and This indicates the speed commands issued to the end effector of the mobile robotic arm and the mobile platform. This represents the pseudo-inverse of the Jacobian matrix corresponding to the main task. Indicates the expected speed of the main task. Indicates the expected location of the main task. Indicates the expected speed of the secondary task. Γ1 represents the expected position of the secondary task, Γ2 represents the position feedback gain coefficient of the primary task, and Γ3 represents the position feedback gain coefficient of the secondary task. (Parameters...) The cross symbol in the upper right corner indicates the calculation of the matrix pseudo-inverse. The relationship between the above tasks and joint velocities can be expressed as follows: Where J1 = J(q,φ), J2 represents the Jacobian matrix of the secondary task, and its pose is expressed as... The Jacobian matrix J2 for obstacle avoidance on the mobile chassis is as follows: J2=(I 3×3 0 3×6 ) Among them, I 3×3 Represents a 3×3 diagonal matrix, 0 3×6 Represents a 3×6 matrix of all zeros; In human-machine collaboration scenarios By integrating the mobile chassis with the additional obstacle avoidance speed v m and the moving chassis speed in the joint speed of collaborative tasks The formula is as follows:

Images