English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present MLNav, a learning-enhanced path planning framework forsafety-critical and resource-limited systems operating in complex environments,such as rovers navigating on Mars. MLNav makes judicious use of machinelearning to enhance the efficiency of path planning while fully respectingsafety constraints. In particular, the dominant computational cost in suchsafety-critical settings is running a model-based safety checker on theproposed paths. Our learned search heuristic can simultaneously predict thefeasibility for all path options in a single run, and the model-based safetychecker is only invoked on the top-scoring paths. We validate in high-fidelitysimulations using both real Martian terrain data collected by the Perseverancerover, as well as a suite of challenging synthetic terrains. Our experimentsshow that: (i) compared to the baseline ENav path planner on board thePerserverance rover, MLNav can provide a significant improvement in multiplekey metrics, such as a 10x reduction in collision checks when navigating realMartian terrains, despite being trained with synthetic terrains; and (ii) MLNavcan successfully navigate highly challenging terrains where the baseline ENavfails to find a feasible path before timing out.

MLNav: Learning to Safely Navigate on Martian Terrains