Implementing a vision based terrain traversability estimator for planetary rovers and astronauts to eliminate heavy hardware used in current techniques. Some of the rovers predict the slip from Wheel odometry but it would be more efficient to be able to predict it in advance if a terrain might be slippery, so as to make more intelligent path planning decisions and to reduce the hardware needed. Classified view of the terrain in front based on their traversability proves to be useful for the rover to take an optimum path and for an astronaut being a helping hand to suggest safest path that could be taken.


Vishnu Teja Yalakuntla


  • Karan Saxena
  • Antonio Del Mastro