The parameters of the Grigoryan soil model are determined using an experimental-computational method previously proposed and the results of reversed experiments on penetration of projectiles with flat and hemispherical heads at impact velocities of 50-450 m/sec in sandy soil. It is shown that the quasistationary dependences of the resistance force on impact velocity obtained in the reversed experiment can be used to solve problems of deep penetration of projectile in soil with an error not exceeding the measurement error.Penetration. The impact and penetration of projectiles in soil have been studied extensively (see, for example, [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]). However, these studies were primarily concerned with the penetration of rigid projectiles in plastic (clay and loamy) soils [8][9][10][11][12][13][14][15][16]. In [13,14], an experimental-theoretical method of dynamic penetration was proposed to determine the shear resistance of clay soil as an ideally plastic medium with the Tresca-Saint Venant plasticity condition; this approach was elaborated in [15][16][17]. Using the hypotheses of incompressibility and ideal plasticity and other simplified concepts of dynamic soil behavior, which are valid to some extent for plastic soil, domestic [1][2][3][4][17][18][19][20][21][22][23][24][25] and [5,6,[26][27][28][29][30][31] foreign authors have developed analytical methods for studying the penetration of rigid bodies of revolution into soil.There have been a few experiments investigating the pulsed loading of loose soil [31][32][33], and the properties of these media are less understood. In experimental studies [34][35][36] of the dynamic compressibility of sandy soils with a plane shock wave, only the shock adiabat of the medium was determined. The compressibility of the medium in penetration experiments was also determined in [37]. The use of a modified Kolsky method [38, 39] with a system of split Hopkinson bar to study the compressibility and plastic properties of soil is restricted by the elastic limit of the material of the measuring bar and holders, which does not exceed 0.5 GPa. Thus, at present, effective methods for studying physicomechanical properties of soil over a wide range of pressure have been developed insufficiently.It seems promising to extend the experimental-computation approach of [40] to study soil properties under dynamic loading using well-known soil models [41,42], methods of mathematical modeling of impact and penetration of deformable projectiles in soil [42][43][44][45][46][47], and data of reversed experiments [10,11]. The dependence of the penetration resistance on the parameters of the soft soil model has been studied previously. It has been established that features of the time dependence of the resistance force allow the force maximum to be used to determine the dynamic compressibility of soil [37], and the quasistationary value can be used to determine its strength (elastoplastic) characteristics [48,49]. A convergent iterative process has been constr...