Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Mission Planning for Autonomous Tree Harvester
In this project we would like to develop a mission planner so that our excavator can autonomously harvest trees in a forest. An algorithm should be developed such that the trees are harvested while minimizing cost of the mission.
Our machine is Menzi Muck walking excavator with walking capabilities and the ability to use out the end-effector tools (http://www.rsl.ethz.ch/robots-media/heap.html). Assuming that the map of the trees to be harvested is given, the machine has to harvest only the labelled trees and leave the rest untouched. In case that some of the trees cannot be reached, the mission planner should come up with a plan that cuts down the minimal number of unlabeled trees.
To achieve this goal, one could use concepts from graph theory, sampling based path planning, dynamic programming etc. The algorithm has to output only the order in which the trees should be harvested. It doesn’t need to consider path planning problem between the individual trees
Bonus: Incorporate traversability data into mission planning.
Our machine is Menzi Muck walking excavator with walking capabilities and the ability to use out the end-effector tools (http://www.rsl.ethz.ch/robots-media/heap.html). Assuming that the map of the trees to be harvested is given, the machine has to harvest only the labelled trees and leave the rest untouched. In case that some of the trees cannot be reached, the mission planner should come up with a plan that cuts down the minimal number of unlabeled trees. To achieve this goal, one could use concepts from graph theory, sampling based path planning, dynamic programming etc. The algorithm has to output only the order in which the trees should be harvested. It doesn’t need to consider path planning problem between the individual trees
Bonus: Incorporate traversability data into mission planning.
- Literature research on optimal planning and graph traversal algorithms
- Problem formulation
- Implementing the most promising concepts and/or implementing your own concepts
- Testing and evaluation
- Literature research on optimal planning and graph traversal algorithms - Problem formulation - Implementing the most promising concepts and/or implementing your own concepts - Testing and evaluation
- Independent working style and good working ethics
- Strong analytical skills and knowledge of discrete optimization
- C++ knowledge and ROS knowledge (optional)
- Independent working style and good working ethics - Strong analytical skills and knowledge of discrete optimization - C++ knowledge and ROS knowledge (optional)
Please contact Edo Jelavic (edo (dot) jelavic (at) mavt (dot) ethz (dot) ch) for any questions. Your application should include a very brief motivational statement, your transcript of records and your CV.
Please contact Edo Jelavic (edo (dot) jelavic (at) mavt (dot) ethz (dot) ch) for any questions. Your application should include a very brief motivational statement, your transcript of records and your CV.