A method for controlling a robot and/or an autonomous driverless transport system on the basis of a sensor-based identification of objects includes generating. Point pair features of the 2D surface contours on the basis of 2D surface contours of the objects to be identified. A point cloud of the environment is acquired using a distance sensor, a surface normal is estimated for each point, and corresponding point pair features of the environment are generated. In a voting method, environment features are compared with model features to efficiently generate pose hypotheses, which are subjected to an optimization and a consistency check in order to ultimately be accepted or rejected as an acquisition.