This study proposes a novel multi-robot navigation algorithm with priority order called, in short, PONA2.0. This algorithm is based on the generalized Voronoi diagram and contains an adjustable multipath switching mechanism and a collision prevention strategy, such that the arrival order of robots is in line with the priority order as much as possible, and the average trajectories length is as short as possible. The given average trajectories length of all robots ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$ATLA$ </tex-math></inline-formula> ) and arrival order ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$AO$ </tex-math></inline-formula> ) are used to be the two performance indices for the comparison between the proposed algorithm and recent existing algorithms NSPP (Huang et al., 2021), PONA (Huang et al., 2022), ROA, and SDA (Ali et al., 2016). The comparison shows that the PONA2.0 can reduce the average <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$AO$ </tex-math></inline-formula> by more than 56% compared with NSPP and PONA and reduce the average <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$ATLA$ </tex-math></inline-formula> between 5% and 17% compared with ROA and SDA.