4 results about "Density evolution" patented technology

The present application relates to a kind of multi-agent collaborative induction and conflict solution method and system based on probability density evolution and wasserstein gradient flow.The method is mapped into continuous probability density field by stratified anisotropic kernel density estimation to the discrete position of multi-agent, constructs control free energy functional in probability space, which fuses target attraction, obstacle repulsion, collision avoidance interaction and Fisher information smoothing regularization term;Based on optimal transport and variational method, derive the macroscopic velocity field that enables the fastest decline of energy, and map into individual control instruction that meets the dynamic constraint.The scheme can improve the safety, smoothness and convergence of cluster flight, and is suitable for unmanned aerial vehicle cluster control and urban low-altitude traffic management.

The present application relates to a kind of vehicle-bridge coupling system reliability analysis method based on adaptive multi-task deep learning, belong to rail transit infrastructure safety technical field.For the problem that existing technology cannot efficiently and synchronously predict vehicle and bridge response, multi-task loss weight adjustment is difficult, adaptive multi-task learning model AMLM is proposed.The model integrates improved gorilla group optimization algorithm MGTO, shared convolutional neural network CNN and double-path state fusion gated long short-term memory network SFLSTM;Through the same variance uncertainty loss weighting strategy, the vehicle and bridge prediction tasks are dynamically balanced;Combined with probability density evolution method PDEM, AMLM-PDEM framework is constructed.The technical effects include: realizing synchronous high-precision prediction of vehicle-bridge response, autonomously optimizing multi-task weight distribution, significantly improving calculation efficiency and modeling accuracy, and perfecting the time-varying reliability evaluation system of vehicle-bridge system.

The invention discloses a channel code pattern optimization method based on Gaussian weighted hybrid approximate density evolution, and relates to the technical field of wireless communication. The method includes constructing a modulation system. A delay module is introduced at a sending end, joint information source and channel decoding is carried out at a receiving end, and likelihood information of delay bits is fed back to a demodulator. Likelihood information probability distribution output by all sub-channels of the system is obtained, and it is confirmed that the distribution does not meet symmetric Gaussian distribution. Based on the characteristics of likelihood information output by each sub-channel under the condition of a single constellation point, decoding convergence performance is analyzed by adopting a density evolutionary algorithm of Gaussian weighted hybrid approximation, and Gaussian weighted hybrid approximation distribution of the likelihood information output by each sub-channel is obtained. And searching and optimizing by using a density evolutionary algorithm based on Gaussian weighted hybrid approximation as an analysis tool to obtain an optimized channel code basis matrix. And constructing a channel encoder by using the optimized channel code basis matrix to replace the original channel code so as to improve the bit error rate performance of the system in the waterfall region.