Dr Abubakr Muhammad Assistant Professor Electrical - - PowerPoint PPT Presentation
EE-562 : Robot Motion Planning Lecture 1 Dr Abubakr Muhammad Assistant Professor Electrical Engineering, LUMS Director, CYPHYNETS Lab http://cyphynets.lums.edu.pk Resources Course material from
Dr Abubakr Muhammad Assistant Professor Electrical Engineering, LUMS Director, CYPHYNETS Lab http://cyphynets.lums.edu.pk
Course material from
http://cyphynets.lums.edu.pk/index.php/Teaching
Textbooks
A robot is an autonomous system which exists in the physical world, senses its environment and acts in it to achieve some goals.
– sensors – effectors/actuators – communication – controller
– acting autonomously – achieving goals
– Electrical engineering (control systems) – Mechanical engineering (mechanisms) – Computer science (AI, learning) – Mechatronics – Bioengineering
– Lawyers (legal issues, labor laws) – Philosophers (ethical issues) – Economists (disruptive technologies) – Social scientists (the social impacts of automation, aesthetics)
Three broad categories
– the philosopher: rationalize the need? – the psychologist: man’s creative traits? – the theologist: man’s obsession with divinity?
– Bible: “God created man in His own image.” (Genesis 1:27) – Hadith: “Allah created Adam in His image.” (Bukhari, Muslim) ﻕﻠﺧ ﷲ ﻰﻠﻋ ﻡﺩﺁﻪﺗﺭﻭﺻ
revolutions – automation
– 1,973 TEPCO emergency workers affected by radiation
– Driving, Mining, Agriculture, Healthcare, Humanitarian, Defense
– Legal challenges – Ethical issues
– Automation and society – Social engineering – Conflicts and responsibility
– Industrial grade manipulators ~ > $100,000 – Baxter (rethink robotics) costs $22,000
safety critical applications
– 4 million driving jobs will be lost
– Who is driving the car? – Accident and insurance – Passenger safety
– Similar issues – Patient care, liability
– Who pulled the trigger? – Wars and international law
– Devolution of governance – Ensuring rights – Conflict resolution
robotics paradigm
– Natural resources – Food and Agriculture Problems – Critical infra-structures – Security
Participation Accountability Entitlements
High Level Specifications Low Level descriptions
Motion Planner
High Level Specifications Low Level descriptions
Motion Planner
High Level Specifications Low Level descriptions
Do this while
Motion Planner
– Tool: Differential equations on continuous spaces – Robustness, stability, optimality – Obstacle avoidance via explicit representations – Model constraints as holonomic / nonholonomic
Related Courses at LUMS CMPE-633B. Robot Dynamics and Control (Fall 2011) CMPE-633C. Geometric Mechanics (Spring 2012)
– Solving puzzles on a discrete space – Focus on algorithmic methods – Obstacle avoidance via explicit labeling
Related Courses at LUMS CMPE 633A. Robot Motion Planning (Fall 2010) CASM Workshop at LUMS Topology of Hyperplane Arrangements (Fall 2012) HEC grant in Mathematics Self Assembly of Hexagonal Structures (2013-14)
– We focus on algorithmic and computational issues in motion planning. – We avoid talking about hardware implementations, software architectures, formal algorithmic complexity analysis.
Related Courses at LUMS EE 622. Applied State Estimation (Spring 2013) ME 410. Probabilistic Robotics (Summer 2011)
– E.g. position & velocity of a robot – Special states: initial state, final state
– A sequence describing chronology of events – E.g. to describe initial, final, intermediate
– Manipulation of the state to get to another state – Inputs or controls
– Strategy of actions acting on states. – E.g. When we are in Initial state move forward.
– Degrees of freedom are high – Structure of configuration / state space is complex – Difficulty in understanding CS topology – Uncertainty in representations (noise)
[Canny 1998].
– Motivation & mathematical representation – Configuration spaces
– No noise – Exact geometry available
– No noise – Practical efficiency vs completeness
– Uncertainties due to sensor and process noise – Real applications
– Open source hands-on tools (ROS, OMPL)