44results about How to "Guaranteed search efficiency" patented technology

The invention discloses a two-stage particle swarm optimization algorithm including independent global search; it includes the steps of initializing the population; adopting a chaotic method to initialize the position X and velocity V of the particles; using the fitness function fitness to calculate the current values ​​of all particles Adaptation value, initialize the historical optimal position pbesti of each particle and the global optimal position gbest of all particles; perform the first-stage iteration-global search; perform the second-stage iteration-local search. Beneficial effects of the present invention: in each iteration of the first stage iteration-global search, the particle randomly selects a non-self particle to learn from all the particles, and this random selection ensures that the group will not track specific particles, and also This ensures that aggregation does not occur. The second stage of iterative-local search can quickly converge and obtain a high-precision solution, the precision of the optimal solution is improved, and the "premature" defect is significantly improved.

A method for planning and dynamic updating of an underwater target search path coordinated by an ROV and a mother ship, comprising the following steps: calculating a path point of the surface mother ship according to a provided search area, and planning the motion path of the surface mother ship; the surface mother ship moves to the way point; Under the condition that the relative relationship between the ROV and the surface mother ship is normal, the current search sub-area is rasterized for the first time. The first layer of grid is the ROV global path planning layer. The biological excitation neural network algorithm is used for the first layer of grid. Carry out global path planning and dynamic update; the first layer of grid is rasterized for the second time, and the second layer of grid is the ROV path dynamic update layer, which realizes the local trajectory planning and dynamic update of the ROV, and judges whether the ROV is triggered in real time. The second-layer grid dynamic update condition is set until the current first-layer grid search is completed. This design achieves full coverage of the search area by forward-looking sonar while ensuring the safe operation and search efficiency of ROV.