In this chapter, we progress beyond the basics of rover vision to introduce some more advanced approaches to vision processing as it underlies the development of greater enhancements to rover missions than any other facet. Often, these require considerable computational effort which may be traded with dedicated hardware. Neuromorphic vision sensors are biomimetic chips that process images rapidly at the focal plane in parallel [511]. For instance, such an approach has been proposed for planetary landers [512]. Alternatively, a log-polar vision sensor with decreasing resolution towards the periphery may be employed to implement active vision though this is unlikely in the context of planetary exploration (this is relatively easily done in software) [513]. The most likely eventuality is that computational resources onboard planetary rovers will evolve greater capacities to enable these greater performances.
CITATION STYLE
Ellery, A. (2016). Rover vision—advanced capabilities. In Planetary Rovers (pp. 263–300). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-03259-2_7
Mendeley helps you to discover research relevant for your work.