A high quality crop seeding operation should enable a rapid and uniform establishment of a desired plant population. Therefore, a no-till seeder must provide a seeding environment that allows the absorption of water by seeds and appropriate temperature and aeration conditions for germination and emergence processes. To stimulate these processes, the seed needs full contact with soil in order to accelerate the absorption of water and oxygen. Covering the furrow with straw is another important aspect, since it prevents soil water losses and surface crusting besides reducing soil thermal amplitude and seedlings stress. Soil contact no-till seeder components are responsible for the accomplishment of this overall function, which depends on correct setting of each individual component for residue cutting, furrow opening, distribution of fertilizer and seeds, furrow closing with soil and straw (grounding) and soil compaction laterally or over the seed. However, the setup and consequently the performance of these components is directly influenced by soil water content, compaction state and particle size of topsoil as well as the amount of straw over soil surface. These factors present a remarkable spatial variability and determine, for the same field, different operational conditions for grounding and compaction components of the seeder. Using of control systems at variable rates with electronic maps is not the best option to solve this problem because the required previous analysis of soil parameters is a time consuming and expensive activity and usually is physical and economically unfeasible. Thus, the quality of no-till seeding can be improved by the development of a real-time control system, operating on-the-go, and controlling the operational parameters of the grounding and compaction components based on information generated by a sensing system of soil conditions. The real-time control system discussed in this paper employs a set of transducers for measurement of soil and operational parameters and based on this information and on computational models estimates the main soil conditions and acts on the components of grounding and compaction in order to get a high seeding quality. The logic of the control system and its functional and non-functional specifications were defined from literature and analysis of field experiments at the Agricultural Research Institute of Paraná State (IAPAR), Brazil, between 2011 and 2012. The system will perform the control, in real time, of two no-till seeder components, i.e. the angle of a pair of grounding discs and the compression of a spring of compaction wheels. The control algorithm employs the autoregressive error function (AREF), a neural network and two arrays. The AREF and the neural network estimate soil moisture from time series of forces acting on a narrow tine and from operation speed. The soil moisture and depth of operation data are the inputs to an algorithm which determines the operational parameters of the grounding and compaction components using two arrays. The neural network with better performance was a Multi-Layer Perceptron type, 2-6-1, with sigmoid activation functions. The AREF uses the time series of forces on the narrow tine with a sample rate of 100 Hz, computing the last 3 m (118 in) readings from sampling position. The control system is under development and will consist of four sensors: a pair of load cells that measure horizontal and vertical forces acting on the tine; a radar that measures the tool operation speed and a laser distance sensor that measures the tool operation depth. The processing will be performed by an ARM microcontroller under a Linux operational platform. Two linear actuators with servo-motors will control the seeder components adjusting the working angle of a pair of grounding discs and the working pressure of the furrow compaction wheel.