That although we have only seen up close and visited a few asteroids and comets they are mostly shaped like 'rubber ducks'?
The 2 joined lobs making what science calls contact binary comets or contact binary asteroids.
Halley's Comet (1P), Comet Borrelly (19P), Comet Churyumov–Gerasimenko (67P), Comet Hartley 2 (103P) are of the rubber duck shape.
Scientists say that to collide and stick together, the impact must have been a very low velocity one - perhaps just a few metres per second. Any faster, and the primary objects would have done enormous damage to each other.
... Other comets seen at relatively close quarters also display lobed shapes, including comets Halley, Borrelly and Hartley-2.
Rosetta's 'rubber duck' comet was once two objects
Could they have been formed or shaped by the same process?
Contact binary asteroids?
Observed near Earth objects have also found to be of the similar contact binary shape or formation - 2063 Bacchus, 4450 Pan, 4486 Mithra, 4769 Castalia, 11066 Sigurd, (179806) 2002 TD66, 2007 TU24, 8P/Tuttle, 2014 HQ124
Contact binary objects?
Are asteroid and comet contact binaries due to an amazingly slow collision among moving masses of dirty ice balls that somehow fuses them together in the freezing cold of space? That comet contact binaries and asteroid contact binaries have happened multiple times?
Unless the reason they are in our orbit is that 2 asteroids/comets have collided and this fusion collision has sent them into our orbits. As comets that we see should only have a relatively short life time as they orbit and sublimate around the sun.
Contact binary planets and moons?
Arecibo radar imagery of Comet 8P/Tuttle reveals a 10-km-long nucleus with a highly bifurcated shape consistent with a contact binary.
... The most important finding from our Tuttle observations is the identification of an apparent contact-binary comet. Recent years have seen a proliferation of discoveries of binary objects among various minor-planet populations, including the near-Earth asteroids (NEAs), main-belt asteroids, Jupiter Trojans, Centaurs, and Kuiper Belt Objects (KBOs). Contact binaries alone are estimated to constitute 10–20% of all NEAs, Trojans, and KBOs. Since comets may be subject to some of the same dynamical processes and/or derive from the same primordial disk population as these other bodies, it has been argued that binary comets should exist and may even be common. Testing this hypothesis has been difficult, however, owing to the sparseness of the sample of comet nuclei that have been studied in detail. Prior to 2008 only six comet nuclei had been imaged by spacecraft or radar. Spacecraft images of Comet 19P/Borrelly revealed a peanut shaped nucleus that has been suggested as a possible contact binary. Comet 1P/Halley is also sometimes described as peanut-shaped and certainly shows some bilobate structure. Although neither of these comets could be considered to be strongly bifurcated, their complex elongate shapes raised the suggestion that some comet nuclei may be composed of coalesced fragments. Tuttle appears to be more strongly bifurcated than either Borrelly or Halley and offers the best evidence yet for the existence of contact-binary comets.
Binary and contact-binary minor planets are thought to form from collisions, mutual capture, or fission.
Radar observations of 8P/Tuttle: A contact-binary comet | Link to PDF