1: \begin{abstract}

2: We consider random walks and L{\'e}vy processes in the free nilpotent Lie group $G^N(\R^d)$ as rough paths. For any $p > 1$, we completely characterise (almost) all L{\'e}vy processes whose sample paths have finite $p$-variation, provide a L{\'e}vy-Khintchine formula for the characteristic function of the signature of a L{\'e}vy process,

3: %treated as a rough path,

4: and give sufficient conditions under which a sequence of random walks in $G^N(\R^d)$ converges weakly to a L{\'e}vy process in rough path topologies. At the heart of our analysis is a criterion for tightness of $p$-variation for a collection of c{\`a}dl{\`a}g strong Markov processes. We demonstrate applications of our results to weak convergence of stochastic flows.

5: \end{abstract}

6: