Hybrid-driven Gaussian process online learning for highly maneuvering multi-target tracking

The performance of existing maneuvering target tracking methods for highly maneuvering targets in cluttered environments is unsatisfactory.This paper proposes a hybrid-driven approach for tracking multiple highly maneuvering targets,leveraging the advantages of both data-driven and model-based algorithms.The time-varying constant velocity model is integrated into the Gaussian process(GP)of online learning to improve the performance of GP prediction.This integration is further combined with a generalized probabilistic data association algorithm to realize multi-target tracking.Through the simulations,it has been demonstrated that the hybrid-driven approach exhibits significant performance improvements in comparison with widely used algorithms such as the interactive multi-model method and the data-driven GP motion tracker.

Dynamic study of infiltration rate for soils with varying degrees of degradation by water erosion

Ultisols,widely distributed in tropical and subtropical areas of south China,are suffering from serious water erosion,however,slope hydrological process for Ultisols under different erosional degradation levels in field condition has been scarcely investigated.Field rainfall simulation at two rainfall intensities (120 and 60 mm/h) were performed on pre-wetted Ultisols with four erosion degrees (non,moderate,severe and very-severe),and the hydrological processes of these soils were determined.The variation of soil infiltration was contributed by the interaction of erosion degree and rainfall intensity (p < 0.05).In most cases,time to incipient runoff,the decay coefficient,steady state infiltration rate,and their variability were larger at the high rainfall intensity,accelerating by the increasing erosion severity.Despite rainfall intensity,the infiltration process of Ultisols was also significantly influenced by mean weight diameter of aggregates at the field moisture content,soil organic carbon and particle size distribution (R2 > 30%,p < 0.05).The temporal erodibility of surface soil and soil detachment rate were significantly and negatively correlated with infiltration rate (r <-0.32,p < 0.05),but less significant correlation was observed between sediment concentration and infiltration rate for most soils,especially at the high rainfall intensity.The variation of surface texture and soil compactness generated by erosion degradation was the intrinsic predominant factors for the change of infiltration process of Ultisols.The obtained results will facilitate the understanding of hydrological process for degraded lands,and provide useful knowledge in managing crop irrigation and soil erosion.

Relationship between granitic soil particle-size distribution and shrinkage properties based on multifractal method

The mechanical properties of granitic residual soils vary with depth due to changes in soil type and heterogeneity caused by weathering.The purpose of this study was to relate the spatial variation of particle-size distribution(PSD)of granitic soils with soil shrinkage parameters using multifractal theory.The heterogeneity of PSD and pedogenic processes were depicted in detail by multifractal dimensions.The PSD generally increased with the increase of profile depth in accordance with the variation of single fractal dimension(D)ranging from 2.45 to 2.65.The shrinkage limit was greatly influenced by the multifractal dimension parameters,including information dimension(D1)and capacity dimension(D0)(Adjusted R2=0.998,P<0.01),and the maximum linear extensibility(κv)was determined by spectral width(?α)and bulk density,with the latter explaining 89%of the total variance ofκv(P<0.01).Soil shrinkage characteristic curve was fitted by the modified logistic model(R2>0.97,root sum of squares<0.1),and the water variation corresponding to the maximum change rate of linear extensibility was determined by the silt content(R2=0.81,P<0.01).Overall,the shrinkage of granitic soils was primarily influenced by PSD and soil compactness.

Generalized labeled multi-Bernoulli filter with signal features of unknown emitters

A novel algorithm that combines the generalized labeled multi-Bernoulli(GLMB) filter with signal features of the unknown emitter is proposed in this paper. In complex electromagnetic environments, emitter features(EFs) are often unknown and time-varying. Aiming at the unknown feature problem, we propose a method for identifying EFs based on dynamic clustering of data fields. Because EFs are time-varying and the probability distribution is unknown, an improved fuzzy C-means algorithm is proposed to calculate the correlation coefficients between the target and measurements, to approximate the EF likelihood function. On this basis, the EF likelihood function is integrated into the recursive GLMB filter process to obtain the new prediction and update equations.Simulation results show that the proposed method can improve the tracking performance of multiple targets,especially in heavy clutter environments.