Lecture

• robot = mechatronic system with perception, decision and action capabilities, that can perform in an autonomous way different tasks in the real world
  • multi-purpose
  • autonomous
• sensor = device that measures a physical property and returns a signal of varying complexity
  • exteroception – information from the outside world
  • interoception – information from within the body of a robot
  • proprioception – information from within related to the movement of the body parts
• actuator = device generating a movement, that controls one degree of freedom (rotation, translation)
  • LED is not an actuator :(
  • effector … more complex for manipulation or mobility
• the three Ds: dull, dirty, dangerous (tasks)
• odometric drift – actuators and sensors are not precise, the error accumulates
  • how to solve it
    • use a map (purposes of a map: localization, computing the route to a given goal)
• motion planning
  • feasible plan = takes the capabilities of the robot into account
  • motion strategies can have different forms – it can be a path, a set of procedural instructions
  • for a walking humanoid robot, the plan would be quite complex
• motion planning vs. navigation (obstacle avoidance etc.)
  • motion planning – global map, strategic level
  • obstacle avoidance – local map, tactical level
• workspace
  • physical space where the robot lives, usually $\mathbb R^2$ or $\mathbb R^3$
  • obstacles, free space
  • types of models
    • continuous metric model
      • polygons – their vertices
      • memory complexity proportional to the number of obstacles
    • discrete metric model
      • pixels: free, fully occupied, partially occupied
      • memory complexity proportional to the size of workspace and the resolution
    • topological model
      • nodes and edges
      • we don't care about the geometry – we only capture places and ways to move from one to another
      • the robot needs to be able to localize itself (what is the current “place”) and to get from one place to another according to the edges in the graph
    • hybrid
  • selection criteria: sensors available, precision required, complexity (memory requirements)
• Piano Mover's Problem
  • free flying robot (rigid object) = piano
  • stationary obstacles (fixed rigid objects)
  • the focus is on the geometry, path planning
• with moving obstacles, time becomes very important
  • we need to know (model) the future motions of the obstacles in order to plan the path
• what is the robot is no longer free flying?
  • two classes of kinematic constraints
    • holonomic – restrict the set of possible position (like obstacles)
      • example: limitations of the robotic arm due to its joints etc.
      • to account for the constraints, we only add some virtual obstacles
    • nonholonomic – restrict the set of possible differential motions
      • example: a wheel rolling without slipping on the plane (you should move in a direction perpendicular to the axis of the wheel)
• prehensile manipulation – we grasp an object and manipulate with it
• non-prehensile manipulation – we push or throw an object
• flexible objects → mechanics of deformation
• uncertainty
  • sources: sensors, actuators
  • model error, action error
• human-robot motion
  • main issues
    • safety – we don't want the robot to collide with people
    • acceptability (politeness) – the robot should not interrupt people engaging in a conversation
  • attention-based HRM
    • we don't want to distract people in the museum
• configuration space
• how to address motion planning
  • mobile robot
    • reactive strategy
    • graph search, A*
    • reinforcement learning
    • robot position … coordinates
  • arm robot
    • pose … vector of joint angles