ANT offers a system (EEGO is the name I believe). Also, there are some wireless systems too from some folks I know. I suppose it depends a bit on how many channels of data you want and also what kind(s) of movements the participants will be making.
g.Nautilus seems to be good choice, and I agree with previous suggestions. However, for me, I believe that the most important is consider having a strong fixed EEG cap with reasonable number of electrodes (e.g., 64 or above ). In addition, you need to pay attention at preprocessing stage as you'll have many temporal muscle activities.
you might want to have a look at the work of Oliveira and colleagues. Besides testing different setups they also published on how to isolate motion artifacts induced into EEG.
Best
Jonas
Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements. Oliveira et al 2016
http://www.ncbi.nlm.nih.gov/pubmed/27242467
Induction and separation of motion artifacts in EEG data using a mobile phantom head device. Oliveira et al 2016