We demonstrate new results using our Spectral LADAR prototype, which highlight the benefits of this sensor for
Unmanned Ground Vehicle (UGV) navigation applications. This sensor is an augmentation of conventional LADAR
and uses a polychromatic source to obtain range-resolved 3D spectral point clouds. These point cloud images can be
used to identify objects based on combined spatial and spectral features in three dimensions and at long standoff range.
The Spectral LADAR transmits nanosecond supercontinuum pulses generated in a photonic crystal fiber. Backscatter
from distant targets is dispersed into 25 spectral bands, where each spectral band is independently range resolved with
multiple return pulse recognition. Our new results show that Spectral LADAR can spectrally differentiate hazardous
terrain (mud) from favorable driving surfaces (dry ground). This is a critical capability, since in UGV contexts mud is
potentially hazardous, requires modified vehicle dynamics, and is difficult to identify based on 3D spatial signatures.
Additionally, we demonstrate the benefits of range resolved spectral imaging, where highly cluttered 3D images of
scenes (e.g. containing camouflage, foliage) are spectrally unmixed by range separation and segmented accordingly.
Spectral LADAR can achieve this unambiguously and without the need for stereo correspondence, sub-pixel detection
algorithms, or multi-sensor registration and data fusion.