Trajectory optimization method and device for accurately deploying marine sensors under water

A trajectory optimization method for accurately deploying marine sensors under water, comprising the following steps:

• step 1:
• according to a preset search range xmin≤x≤xmax, N sets of values are randomly selected for initial control variables x=(U0, V0, W0, ωx0, ωy0, ωz0, ϕ0, θ0, ψ0), denoted as xi, wherein, i=1, 2 . . . , N, representing an i-th set, xmin, xmax are a lower limit and an upper limit of the search range, U0, V0 and W0 are components of an initial linear velocity in x-, y-, z-directions in an local coordinate system, ωx0, ωy0 and ωz0 are components of an initial angular velocity ω0 in the x-, y-, z-directions in the local coordinate system, and (ϕ0, θ0, ψ0) are initial Euler angles of the sensor;
• step 2:
• in an n-th iterative operation, all N sets of xi are input into an underwater glide kinematics and dynamics model for an underwater trajectory of the sensor, and N actual deployment positions are calculated, then, a minimum deployment error Δs(xi)nmin between a target deployment position and the actual deployment position of each set of xi in the current iterative operation is calculated, thereafter, minimum distances corresponding to the N sets of xi are compared with each other, a minimum value is taken as a global minimum deployment error Δs(x)nminmin of this iterative operation, wherein:
• when n=1, Δs(xi)lmin=Δs(xi)1, Δs(x)1minmin is denoted as follows, and turn to step 3:
• Δs(x)1minmin=Min{Δs(x1)1, Δs(x2)1, . . . Δs(xN)1};
• when n>1, Δs(xi)nmin=Min{Δs(xi)n, Δs(xi)(n-1)min}, and then turn to step 3, under the circumstances,
• Δs(x)nminmin=Min{Δs(x1)nmin, Δs(x2)nmin, . . . Δs(xN)nmin}
• wherein a subscript n represents current iteration operation steps, Δs(xi)n represents an error distance between the actual deployment position and the target deployment position in the n-th step of iterative operation in the i-th set, a subscript min is a mark for a minimum value in the set, and a subscript minmin is a mark for minimum value between the sets, Δs(xi)nmin represents the internal minimum deployment error in the n-th step of iterative operation in the i-th set, and Δs(x)nminmin represents a global minimum deployment error of a whole domain in the n-th step of iterative operation;
• step 3:
• determine whether n is less than a preset maximum number of iteration steps N0, if yes, a global random walk is performed, N sets of xi are updated, and then a local random walk is performed on the above updated N sets of xi according to a preset probability, the N sets of x are obtained again, and return to step 2;
• otherwise, go to step 4;
• step 4:
• the control variable xi corresponding to Δs(x)nminmin and a corresponding trajectory are output as an optimal control variable and an optimal trajectory,
• wherein the optimal control variable and the optimal trajectory are used to improve an accuracy of a three-dimensional underwater trajectory prediction for the marine sensor to be deployed underwater.