No, no!
Stanley, what do you want to detect with your ultrasonic detector, the biggest echo or the first? The biggest echo will result from the largest reflection plane, while the first echo will result from the nearest object. So, if you want to detect the nearest and very probably most 'dangerous' object next to your robot, then you need to detect the first echo, which results from a direct signal path from your transmitter to the object and back. This will always be the first echo. (If you have your detector close to a plane from which a parastic echo originates, then you can suppress this echo either by a horn arrangement on the transducers or by cancelling via software. But nevertheless, what you want to detect is the first echo, not the biggest.)

The reason, why you think that the first echo is wrong in this plot, which I repeat here



is, that you give the pulses not enough time between them. So, what you see as first echo (named 'false echo' in your plot) actually is the third, fourth or fith echo resulting from some much earlier emitted pulse! This isn't the first echo of the pulse you assume!!
Stand in your bathroom and clap your hands. Estimate how long the reverberation (echoes which are so many that they smear) lasts? This time can be many many 100msec! So, once having emitted your ultrasonic pulse you must wait long enough before emitting the next one. Otherwise you will suffer from late reflections, eroneously be detected as first echoes.

Kai