Perhaps expanded NEO (near earth objects) program for larger(smaller?) objects for very near approach to earth. Perhaps a lunar orbiter with fish-eye camera looking for optical flashes and motion, and an infrared camera for further motion detection? Also place cameras at Lagrange point(s)? Better yet, cameras at 4 Lagrange points allowing (for at least 2 sightings) triangulation determination of trajectory, sufficient enough to determine if such object is a re-visit, or new to database. Yet size and mass of such objects would still be undetermined; unless fish-eye camera (?) at geosynchronous altitude (~25k?), and for lunar orbiter? So does our current database of NEO include a number of revisits of same objects? Might not yet arriving intersecting objects’ trajectories be detected, rather than just after the fact encounters? Hence perhaps a different database of objects for such encounters, and any additional risk i.e. adjusted odds? Might the recent Russian meteorite encounter be consistent with a not yet revealed different (larger or smaller) database? Might lights seen by pilots high in sky represent fireballs occurring at higher altitude; thus changing the odds?
Might not all of mass from planetesimal collision with proto-earth 4,5 byrs ago have gone into low orbit and congealed into our then low orbit moon? Even if just a smaller percentage escaped, it could still represent a large mass amount and large data set of objects, in analogy to Oort cloud. Would such material from earth and planetesimal collision seem to be of mantle origin, and hence of a rock/stoney nature? Wouldn’t such debris go into solar orbit, not so dissimilar in length from earth’s; hence on repetitive intercepting trajectories? Would this change the odds for collisions for all sizes, inclusive of much smaller than any mentioned .6 mile diameter? Even considered size of 100 meters or smaller might seem of significance, if part of a much larger data set. So have we drastically mis-estimated the odds of significant collision, in part related to unknown size and unknown trajectory; and also in part failing to appreciate the consequences of a planetesimal-earth collision 4.5 byrs ago? Another imprnit of such ancient event?
How might one test such conjecture, that we are still exposed to consequences of a collision 4.5 byrs ago? On Mars, ground penetrating radar has indicated a sub-surface ice-like reflections. Did it also show more definitive evidence of significant larger cratering? If not, then the Late Heavy Bombardment (LHB) would not have seemed to be a solar system wide event. Rather such LHB was then just likely peculiar to our earth-moon system; hence consistent with the LHB representing return of fragmentation from such proto-earth and planetesimal collision 4.5 byrs ago. Most larger pieces, since probably less abundant, might have been cleared; but likewise for all smaller pieces (however defined)? Is thus the risk greater than we surmise; or at least is the database quite different (even smaller?) from what we currently have? Again, have we mis-estimated the odds for return home of any such fragmentation? TMM