Considering the uncertainty of measurement origin deriving from the random surviving multi-target, clutter, and the missed detection of sensor, a multi-range hypothesis Bernoulli filter is proposed in order to solve the bearings-only tracking problem. Firstly, a multi-range hypothesis based bearings-only measurement equation is modeled in Cartesian coordinates. Then, a Gaussian mixture multi-range hypothesis cardinality-balanced multi-target multi-Bernoulli filter is derived and proposed. Finally, an adaptive bearings-only birth multi-Bernoulli density is proposed. The simulation results show the effectiveness of the proposed approach.