Robot inverse kinematics solution prediction method based on POA-transformer-BiGRU-attention integrated optimization algorithm

A robot inverse kinematics solution prediction method based on a POA-Transformer-BiGRU-Attention integrated optimization algorithm, wherein the method comprises:

• initializing a POA by randomly generating individual robot positions Pi;
• wherein the generating individual robot positions Pi comprises:
• setting a number of the robot as N and a maximum iteration counts as Tmax;
• initializing individual robot positions as:
• Pi={xi1,xi2, . . . xid,}(i=1,2, . . . , N), where xid represents a value of a i-th data point in a d-th feature dimension;
• evaluating and updating the individual robot positions Pi using a fitness function to obtain updated positions
• P
• i
• new
• ;
• defining an input data
• X
• i
• =
• P
• i
• new
• ,
• then training the Xi through a Transformer model to generate output features Z′;
• processing the output features Z′ using a BiGRU to obtain final hidden states, and applying an attention mechanism to compute weighted summation results z;
• and
• generating predicted inverse kinematics solutions based on the weighted summation results z, evaluating an accuracy of the inverse kinematics solutions using an objective function, and determining a global optimal solution;
• wherein the evaluating an accuracy of the inverse kinematics solutions using an objective function comprises:
• calculating the objective function f(θ) as:
• f(θ)=ω1·Eposition(θ)+ω2·Eorientation(θ)+ω3·C(θ)+ω4·Eprediction(θ);
• where θ is a joint angle vector, ∫1, ω2, ω3, ω4 are weighting coefficients, Eposition is an position error;
• Eorientation is an orientation error;
• Eprediction is a prediction error, C(θ) is a joint limit constraint;
• wherein the position error is calculated as:
• Eposition(θ)=∥Pdesired−Pactual(θ)∥2;
• where Pdesired is a desired end-effector position vector;
• Pactual(θ) is an actual end-effector position vector;
• wherein the orientation error is calculated as:
• Eorientation(θ)=∥Rdesired−Ractual(θ)∥F;
• where Rdesired is a desired end-effector rotation matrix;
• Ractual(θ) is an actual end-effector rotation matrix;
• wherein the joint limit constraint is calculated as:
• C
• ⁡
• (
• θ
• )
• =
• ∑
• i
• =
• 1
• 6
• [
• penalty
• ·
• max
• (
• 0
• ,
• θ
• i
• -
• θ
• i
• ,
• m
• ⁢
• a
• ⁢
• x
• )
• 2
• +
• max
• ⁡
• (
• 0
• ,
• θ
• i
• ,
• mi
• ⁢
• n
• -
• θ
• i
• )
• 2
• ]
• ;
• where penalty is constraint coefficient, θi is a i-th allowable rotation angle;
• θi,max is a maximum allowable rotation angle for a i-th parameter;
• θi,min is a minimum allowable rotation angle for a i-th parameter, wherein if θi exceeds a defined robotic arm range θi,limit, a penalty term is applied to enforce convergence to valid search space;
• minimizing a value of the objective function f(θ) via iterative parameter adjustment and POA optimization to ensure the accuracy of the inverse kinematics solutions and obtain the global optimum solution.