Dynamical mass estimates for the main asteroid belt and the trans-Neptunian Kuiper belt have been found from their gravitational influence on the motion of planets. Discrete rotating models consisting of moving material points have been used to model the total attraction from small or as yet undetected bodies of the belts. The masses of the model belts have been included in the set of parameters being refined and determined and have been obtained by processing more than 800 thousand modern positional observations of planets and spacecraft. We have processed the observations and determined the parameters based on the new EPM2017 version of the IAA RAS planetary ephemerides. The large observed radial extent of the belts (more than 1.2 au for the main belt and more than 8 au for the Kuiper belt) and the concentration of bodies in the Kuiper belt at a distance of about 44 au found from observations have been taken into account in the discrete models. We have also used individual mass estimates for large bodies of the belts as well as for objects that spacecraft have approached and for bodies with satellites. Our mass estimate for the main asteroid belt is $(4.008 \pm 0.029)\cdot 10^{-4} \ m_{\oplus}$ (3$\sigma$). The bulk of the Kuiper belt objects are in the ring zone from 39.4 to 47.8 AU. The estimate of its total mass together with the mass of the 31 largest trans-Neptunian Kuiper belt objects is $(1.97 \pm 0.030)\cdot 10^{-2} \ m_{\oplus}$ (3$\sigma$), which exceeds the mass of the main asteroid belt almost by a factor of 50. The mass of the 31 largest trans-Neptunian objects (TNOs) is only about 40% of the total one.

The Kuiper belt includes tens of thousands of large bodies and millions of smaller objects. The main part of the belt objects is located in the annular zone between 39.4 au and 47.8 au from the Sun, the boundaries correspond to the average distances for orbital resonances 3:2 and 2:1 with the motion of Neptune. One-dimensional, two-dimensional, and discrete rings to model the total gravitational attraction of numerous belt objects are considered. The discrete rotating model most correctly reflects the real interaction of bodies in the Solar system. The masses of the model rings were determined within EPM2017---the new version of ephemerides of planets and the Moon at IAA RAS---by fitting spacecraft ranging observations. The total mass of the Kuiper belt was calculated as the sum of the masses of the 31 largest trans-neptunian objects directly included in the simultaneous integration and the estimated mass of the model of the discrete ring of TNO. The total mass is $(1.97 \pm 0.30)\cdot 10^{-2} \ m_{\oplus}$. The gravitational influence of the Kuiper belt on Jupiter, Saturn, Uranus and Neptune exceeds at times the attraction of the hypothetical 9th planet with a mass of $\sim 10 \ m_{\oplus}$ at the distances assumed for it. It is necessary to take into account the gravitational influence of the Kuiper belt when processing observations, and only then to investigate residual discrepancies to discover a possible influence of a distant large planet.

We have searched for and estimated the possible gravitational influence of dark matter in the Solar system based on the EPM2011 planetary ephemerides using about 677 thousand positional observations of planets and spacecraft. Most of the observations belong to present-day ranging measurements. Our estimates of the dark matter density and mass at various distances from the Sun are generally overridden by their errors ($\sigma$). This suggests that the density of dark matter ${\rho}_{dm}$, if present, is very low and is much less than the currently achieved error of these parameters. We have found that ${\rho}_{dm}$ is less than $1.1\cdot10^{-20}$ g cm$^{-3}$ at the orbital distance of Saturn, ${\rho}_{dm}<1.4\cdot10^{-20}$ g cm$^{-3}$ at the orbital distance of Mars, and ${\rho}_{dm}<1.4\cdot10^{-19}$ g cm$^{-3}$ at the orbital distance of the Earth. We also have considered the case of a possible concentration of dark matter to the Solar system center. The dark matter mass in the sphere within Saturn's orbit should be less than $1.7\cdot10^{-10} M_{\odot}$ even if its possible concentration is taken into account.

The high precision of the latest version of the planetary ephemeris EPM2011 enables one to explore more accurately a variety of small effects in the solar system. The processing of about 678 thousand of position observations of planets and spacecrafts for 1913--2011 with the predominance of modern radar measurements resulted in improving the PPN parameters, dynamic oblateness of the Sun, secular variation of the heliocentric gravitational constant $GM_{\odot}$, and variation range of the gravitational constant $G$. This processing made it possible to estimate the potential additional gravitational influence of dark matter on the motion of the solar system bodies. The density of dark matter ${\rho}_{dm}$, if any, turned out to be substantially below the accuracy achieved by the present determination of such parameters. At the distance of the orbit of Saturn the density ${\rho}_{dm}$ is estimated to be under $1.1\cdot10^{-20}$ g/cm$^3$, and the mass of dark matter in the area inside the orbit of Saturn is less than $7.9\cdot10^{-11}$ M$_{\odot}$ even taking into account its possible tendency to concentrate in the center.

More than 635 000 positional observations (mostly radiotechnical) of planets and spacecraft (1961-2010), have been used for estimating possible changes of the gravitation constant, the solar mass, and semi-major axes of planets, as well as the value of the astronomical unit, related to them. The analysis of the observations has been performed on the basis of the EPM2010 ephemerides of IAA RAS in post-newtonian approximation. The obtained results indicate on decrease in the heliocentric gravitation constant per year at the level $ \dot {GM_{Sun}}/GM_{Sun} = (-5.0 \pm 4.1) 10^{-14} (3\sigma).$ The positive secular changes of semi-major axes $ \dot a_i/a_i$ have been obtained simultaneously for the planets Mercury, Venus, Mars, Jupiter, Saturn, as expected if the geliocentric gravitation constant is decreasing in century wise. The change of the mass of the Sun $M_{Sun}$ due to the solar radiation and the solar wind and the matter dropping on the Sun (comets, meteors, asteroids and dust) was estimated. Taking into account the maximal limits of the possible $M_{Sun}$ change, the value $\dot G/G$ falls within the interval $ -4.2\cdot10^{-14} < \dot G/G < +7.5\cdot10^{-14}$ in year with the 95% probability. The astronomical unit (au) is only connected with the geliocentric gravitation constant by its definition. In the future, the connection between $GM_{Sun}$ and au should be fixed at the certain time moment, as it is inconvenient highly to have the changing value of the astronomical unit.