Neat stuff to think about! I have a couple of comments to share and I apologize if I end up sounding like a teacher or something. It's because I just got my PhD in physical anthropology and am really immersed in this topic.
Lots of people study the origins of human locomotion, how/when/where it evolved, how efficient it is, and so on. Just last week in Nature there was an article that showed that running barefoot is more efficient than running with shoes, because you land on the balls of your feet and use your calf muscles to better advantage.
There are a lot of theories out there about the origin of bipedalism. Like Tom said, the consensus is that bipedalism evolved as a way of getting quickly from one small forested area to another, in a time period when the environment was getting less uniform and more "patchy". This was happening well before A. ramidus or A. africanus since those guys already had biology that was pretty well adapted to bipedalism (they probably had a different kind of bipedalism from ours - just like there are different kinds of quadrupedalism: compare, say, a horse, which runs on just the middle toe (the others have disappeared over time) and a bear, which walks on the flat of its foot, like we do).
JP is right that bipedalism evolved well before we were using tools. Walking bipedally would have allowed hominins to travel through areas of thick grass and see over the tops of the grass (and thus able to better see the predators), to travel longer distances (bipedal walking is more efficient than chimp's knucklewalking), and possibly to better thermoregulate (because, if you're standing upright in an open, grassy environment, less of your body is hit by direct rays of the sun).
GoldenSlumbers mentions a recent theory about A. ramidus, that bipedalism evolved as a mechanism for males to provision females. This, unfortunately, is the result of one guy's theory becoming really popular, but it's not backed up by any real evidence. The excavator of the A. ramidus fossils, Owen Lovejoy, suggested that because the male and female A. ramidus had similarly-sized canines, this indicated there was no competition between males for mates. In his mind, this lack of competition led to everything from increased female breast size to nuclear families and male provisioning, even though there's NO EVIDENCE at all for this kind of social organization. Many people in the field think that this is terrible science and are shocked that it got published in Nature, and several, including people in my department, are writing a response to Nature.
As for the environment in which this change occurred, many of the early hominin fossils are found in riverine/forested environments, that are usually surrounded by dry, grassy plains. I'm pretty sure that bipedalism didn't evolve for navigating the wetlands and swamps, especially because rivers and water-sources would have been dangerous places for hominins to be - crocodiles being common, and other predators using water holes as hunting grounds. Instead, the forests around the rivers would have been the preferred habitat, and getting across the plains from one forested area to another was where bipedalism gave an advantage.
Our more ancient ancestors evolved to move around in the complex, 3-D environments of forests, which gives us a lot of benefits, not the least of which is 3d vision! We do have fantastically mobile shoulders - not as much as apes do, for sure, but gibbons and orangs spend more of their time in the trees than chimps - they are amazing to watch. While we're more distantly related, we also share in common some traits with baboons (especially things like finger, hand and foot length ratios). Some baboon species spend a lot of time climbing and sleeping in cliffs. You might say that our morphology, while perhaps not evolved specifically to cliff-climbing, does allow us to climb things other than trees, at least much better than our long-fingered and -toed ancestors did.
Finally, I was intrigued by your comment about meat-eating and the evolution of brain size. That is another area of huge debate, and one near and dear to my heart. There's some new evidence that it's not just eating meat, but eating cooked foods in general (meat AND vegetable) that would have allowed this brain size increase. But that's a bit off topic for here.
yeesh, sorry to ramble on so much! I'm really excited that there are people out there who are thinking of how our body can move outside of how we "usually" do.