Or, you combine rides. Stanford University is working on a standardized "cubesat", 4" on a side. A standard dispenser holds 3 of them. The dispensors are mounted as secondary payloads on rockets (or LOTS of them as primary payloads). I believe that their estimated cost to orbit is $50k for one cubesat.
It's not a matter of one or the other. You send spacecraft to do dull, routine stuff (like handle communications relays) or exploration that is too dangerous or far-away for people. However, robots only do what they're engineered to do, witness the Deep Space 1 probe that totally missed photographing the one asteroid because it couldn't get the camera gain set right. You send people, usually after the robots, to look at things and do on-site decision making. Only a human can say "hmm, that looks wierd, I wonder what will happen if...."
Or, you combine rides. Stanford University is working on a standardized "cubesat", 4" on a side. A standard dispenser holds 3 of them. The dispensors are mounted as secondary payloads on rockets (or LOTS of them as primary payloads). I believe that their estimated cost to orbit is $50k for one cubesat.
It's not a matter of one or the other. You send spacecraft to do dull, routine stuff (like handle communications relays) or exploration that is too dangerous or far-away for people. However, robots only do what they're engineered to do, witness the Deep Space 1 probe that totally missed photographing the one asteroid because it couldn't get the camera gain set right. You send people, usually after the robots, to look at things and do on-site decision making. Only a human can say "hmm, that looks wierd, I wonder what will happen if...."