Lecture 7 - Manipulation and Sensors Flashcards
Manipulation
A goal-driven movement of any type of manipulator (effector)
Manipulators
- What is the name of general components? How many does it have?
- What are they connected by?
- What is each end connected to (2 terms)?
- Manipulator links, several
- Manipulator links are connected by joints
- Robots body and endeffector
How can manipulator joints be controlled?
Independently
What has to be avoided during movement (2)?
- Overextention of joints (joint limits) and other body parts
- Obstacles in environment
Free space / Workspace
The space in which movement is possible and in which a path has to be searched.
2 Types of Joints
1. Rotary (revolute) joint
Rotational movement around a fix axis (human/mammal joints)
2 Types of Joints
2. Prismatic joint
Smaller round metal piece in bigger metal piece that can slide out (slider)
- Does a joint provide a DoF? If yes, which one and how many?
- What does an endeffector with many (c)DoF’s need?
- Yes, normally one cDoF. For each additional one, a seperate actuator is needed
- Many cDoF require many actuators.
“Human like in shape”
Anthropomorphic
Characteristics of a anthropomorphic hand (3)
- Large number of joints
- Large number of DoF
- Compact design
What do we have to know to control a robot manipulator?
Its kinematics
Kinematics
The correspondence between actuator motion and the resulting effector motion
(3) Examples teleoperated robots
- robot-assisted surgery
- Space robots
- Rescue robots/operations
(3) Challenges teleoperated robots
- Complexity of manipulator (DoF’s)
- Constraints of the interface
- Limitations of sensing
Teleoperation can be combined with…
Autonomy! A system that is not continuously controlled
Forward Kinematics
For a robot with n joints, what is the endeffector pose, given the joint angles (q)
Inverse Kinematics
Which joint movements (q) are needed to achieve a particular robot endeffector pose?
-> move endeffector to a desired position
Forward Kinematics - Formula
1. What is a1,2 / x1,y1 / xe, ye/ q1,2?
a1 & 2 = length of links
x1 & y 1 = position of moving joint (other one is at 0,0)
xe & ye = position of endeffector
q1 & 2 = angles of first and moving link
Forward Kinematics - Formula
2. Equation for x1 & y1
x1 = a1cos(q1)
y1 = a1sin(q1)
q1 & 2 = angles of first and moving link
x1 & y 1 = position of moving joint (other one is at 0,0)
Forward Kinematics - Formula
3. Equation for xe & ye
xe = x1 + a2*cos(q1+q2) ye = y1 + a2*sin(q1+q2)
xe & ye = position of endeffector
Simple Kinematic Model
1. Linear Velocity
- Denoted as V, means speed (forwards, backwards)
Simple Kinematic Model
2. Angular velocity
- Denoted as weird W, speed of turn (sideways)
Simple Kinematic Model
Velocity of robot’s center (linear) formula
V = r/2 (Vl + Vr)
r = wheels radius
Vl & Vr = velocity left/right wheel
Simple Kinematic Model
Angular Velocity formula
W = R/d (Vl-Vr) d = distance between wheels r = wheel radius
What does Manipulation involve (3)?
- Trajectory planning
- Kinematics
- Dynamics (properties of motion and energy of a moving object)
Matrix Muliplication
- A(row column) or A(column row)?
- When is MM possible?
- What are the dimensions of the result of a MM?
- A(rows columns)
- MM is possible iff
columns of A == rows of B - A(rA cA) * B(rB cB) = C(rA cB)
Sensor Classification
1. Proprioceptive sensors
Measures values internal to the system (odometer, motor speed, joint angles, battery voltages)
Sensor Classification
2. Exteroceptive sensors
Acquire information from the environment (distance, light intensity, sound amplitude)
Sensor Classification
3. Passive sensors
Measure ambient environmental energy entering the sensor (temperature, microphone)
Sensor Classification
4. Active sensors
Emit energy into the environment and measure the environmental reaction (encoders, laser range finders, ultrasonic sensors)
What are sensor and what can they measure?
They are physical devices that measure physical quantities
Uncertainty in Robotics
1. Definition
- Inability to be certain about the state of the robot
Uncertainty in Robotics
2. (4) sources
- Sensor noise & errors
- Sensor limitations
- Effector and actuator noise and errors
- Lack of knowledge about the environment or dynamic (changing) environment
Properties of Sensors
1. Dynamic range
+ example
- Ratio of the maximum input value to the minimum measurable input value
In photography: ratio (one number) between the max and min measurable light intensities
Properties of Sensors
2. Resolution
Minimum difference between two measured values that can be detected by a sensor
-> often the lower limit of the dynamic range
Properties of Sensors
3. Linearity
The plot of a sensor’s input and output response is a straight line
Properties of Sensors
4. Bandwidth of Frequency
+ in what is it expressed?
+ is higher better?
The number of measurements per second
+ expressed in Hertz
+ yes, always
Sensor information
Signal-to-Symbol problem + solution
The sensor output has to be translated to useful information
-> sensor processing is needed
Simple Sensors
(3) switches
1. Contact sensor
2. Limit sensor
3. Shaft encoders
- Detect contact with another object (grabbing object/hitting wall)
- Detect when a mechanism has moved to the end of its range (when is grabber wide open?)
- Detect times a motor shaft turns, by having a switch click every time the shaft turns
Simple Sensors
Light sensor
1. What are they?
2. What do they detect (3)?
1. They are photocells 2. (1) Light intensity (2) Differential intensity (between sensors) (3) Break in continuity (break beam)
Simple Sensors Reflective Optosensors 1. Parts (2) 2. Reflectance sensors 3. Break beam sensors
- (1) Emitter (LED) and (2) Detector
- Reflectance sensor detects presence of object when the light reflects
- Break beam sensor detects presence of object when light is interrupted
2 Types of joints
- Rotary joint
2. Prismatic joint
Simple Sensors (3) switches
- Contact sensor
- Limit sensor
- Shaft encoders
Thing that emits and senses light
+ two different kinds
Reflective Optosensor
- Reflectance sensor
- Break-beam sensor
