Because machine learning methods are often heuristic, issues such as safety and performance are critical.
Also, learning-based questions such as problem learnability, knowledge transfer among robots, knowledge generalization, long-term autonomy, task formulation, demonstration, role of simulation, and methods for feature selection define problem solvability.
The topics include:

  • Task representation and classification
  • Planning for complex and high dimensional environments
  • Smart sampling techniques for motion planning
  • Learning feature selection
  • Methods for incorporating learning into planning
  • Reinforcement learning for robotics and dynamical systems
  • Transfer of learning and motion plans, knowledge and experience sharing among the agents
  • Policy selection: exploration versus exploitation, methods for safe exploration
  • Methods for creating motion plans that meet dynamical constraints
  • Task planning and learning under uncertainty and disturbance
  • Motion planning for system stability
  • Adaptable heuristics for efficient motion plans
  • Motion generalization – methods that learn subset of motion and produce plans with higher range of motions
  • Motion planning for multi-agent systems and fleets