ABSTRACT In this study, we investigated the optimal placement of ground‐based sensors (or anchors) for three‐dimensional multilateration based on the least‐squares approach. Under an equidistance assumption for the target‐anchor distances, which is reasonable for ground‐based systems, we derive a cost function based on the error variance of the least‐squares solution for optimal anchor placement without relying on prior knowledge of the target location. We propose specific anchor placements for boundary constraint scenarios. The simulation results confirmed that the proposed anchor placement method yielded the optimal average error variance for various target locations.