In whatever size groups, it is clear that ankylosaurs had a very low browsing range, foraging no more than a meter or so above the ground. F. Nopcsa once suggested that ankylosaurs were insectivorous, but this idea is no longer given much credence. Instead, it is now thought that these animals fed exclusively on plants. Why this is so comes from a look at the digestive system, from the beak, mouth, and teeth at the front end, as well as the far extreme of the gut.

At the front end of the ankylosaur, the beak was covered by horn, making this rhamphotheca both strong and sharp. Primitively in ankylosaurs, the beak is scoop-shaped but relatively narrow (and remains so in the nodosaurid clade and in Shamosaurus), but in all other ankylosaurids, the beak becomes very broad, with some forms losing the premaxillary teeth all together. As noted by K. Carpenter of the Denver Museum of Natural History, this in turn could indicate differences in the foraging specificity among the ankylosaurs; nodosaurids being more selective, while others could be interpreted as much more general feeder, foraging on all low-lying vegetation they could find.

Despite having similar skull morphology as stegosaurs; small, unelaborated triangular teeth that are loosely arranged, tooth wear indicates that chewing involved not only puncturing of foods but also grinding action as well. In addition, it is likely that ankylosaurs had a long, flexible tongue and an extensive secondary palate that allowed them to chew and breath at the same time, with well-developed deep cheek pouches to keep whatever food was being chewed from falling out of the mouth. Likewise, the jaw bones themselves were relatively large and strong (despite lacking enlarged areas for muscle attachment.) Many ankylosaur jaw features – except for tooth design and placement – suggest that ankylosaurs were reasonably adept chewers.

The posterior end of the feeding apparatus of the ankloysaurs were similar to that of modern ruminants, such as cows, which use symbiotic bacteria within the stomach (or stomachs, because ankylosaurs may well have had series of them). The stomach would have served as a great fermentation vat, decomposing even the toughest, woody plant material. In ankylosaurs, the very deep rib cage circumscribed an enormously expanded abdominal region and this can only mean that digestion took place in a very large, perhaps highly differentiated, fermentation compartment.

The combination of browsing at low levels and having anatomy indicative of chewing and fermenting places limits on what kinds of plant ankylosaurs may have fed on. At the levels where ankylosaurs concentrated, the undergrowth consisted of a mixture of low-stature ferns, gymnosperms such as cycads and bennettitaleans, and, during the Late Cretaceous, shrubby angiosperms.

References;
Farlow, J. O. (1978) Speculations about the diet and digestive physiology of herbivorous dinosaurs, Paleobiology 13, 60-72.
Fastovsky, D. E. & Weishampel, D. B. (2005) The Evolution and Extinction of the Dinosaurs (2nd Ed.), Cambridge University Press: Cambridge, UK.