Research On Drilling Strategy Of Lunar Drilling And Coring Based On Real-time Recognition Of Drillability

Currently, China is implementing the lunar exploration mission named asChang’e. Due to efficient penetrating and cuttings removal traits, drilling and coringhas been selected as the main sampling method in planetary sampling history.Considering that physical properties of lunar regolith on different sampling sitesand even at different depths on one site are quite different, drilling&coring deviceshould have high adaptability to different drilling media. Apart from mechanicalstructures’ optimization, the most effective way to improve drill tool’s adaptabilityis probably to establish an intelligent drilling strategy.Different with drilling on the earth, lunar drilling&coring method has manydifficulities in penetrating the undetermined drilling targets with limited orbitalsensor resources. Since there are quite a few physical parameters of drilling medium,it is hardly to control the drill tool by recognizing all mechanical parameters oflunar regolith. Hence, this paper proposed the drillability concept to simplify thedescription of drilling difficuilty. By establishing the interaction model betweendrill tool and lunar regolith, reasonable drilling parameters matched with drillabilitylevel can be acquired. Once drilling medium’s drillability has been identified by useof pattern recognition method, corresponding drilling parameters will be turned toadapt to the recognized medium.Firstly, hollow auger and drilling&coring bit are selected as the targets toconduct the research of lunar regolith’s failure mode and conveyance mechanism.Based on soil mechanics methods, the concept of Filling Rate of the Auger (FRA) isproposed to distinguish the drilling conditions. The interaction models between drilltool’s structure parameters, drilling parameters and lunar regolith’s physicalparameters in two drilling drilling conditons have been finally established.Secondly, a drilling and coring test bed’s control system and typical lunarregolith simulants are prepared in advance. Experiments under different drillingparameters are conducted to verify the rationality of FRA and the interaction models.Using the principle of least energy consumption and a few optimization indexes,reasonable drilling parameters in granular lunar regolith have been acquired. Finally, lunar regolith simulants’ drillability indexes have been obtained undermechanical penetrating method. After analyzing drilling signals’s time andfrequency domain traits, a group of characteristic signals are selected as the input ofdrillability recongnition module. By an effective pattern recognition method SVM,an optimized drillabiltiy recognition algorithm has been designed to identify drillingmedium accurately and turn the drilling parameters automatically. Experimentsdemonstrated the feasibility of this real-time recognition based drilling strategy.