Intuitive Programming of AIMM Robot
Authors
Carøe, Christian ; Hvilshøj, Mikkel ; Schou, Casper
Term
4. term
Education
Publication year
2012
Submitted on
2012-06-01
Pages
162
Abstract
This thesis presents the development of a mobile manipulator, Little Helper ++, and an accompanying software system that enables intuitive robot programming for shop-floor operators without robotics expertise. Conducted in collaboration with the EU-funded TAPAS project, the work aims to create a flexible industrial assistant. Little Helper ++ integrates a Neobotix platform, an electric gripper, and a KUKA Light Weight Robot, and is controlled by the Little Helper System—an task-level, layered software architecture with device primitives, skills, and complete tasks as three levels. The system is operated via both a terminal interface (TUI) and an iPad-based graphical user interface (GUI), combined with physical teaching in which the operator guides the robot to learn coordinates. A library of skills, including classic robot macros such as pick and peg-in-hole, was developed—ten skills in total—that can be composed to solve complex tasks. The approach was tested with a non-robotics expert who, after a brief introduction, was able to program and execute an industrial task replicated from Grundfos. The system was further verified by programming and carrying out two complex assembly tasks from Grundfos, indicating that the skill-based, task-level approach effectively supports intuitive programming and flexible automation.
Specialet beskriver udviklingen af en mobil manipulator, Little Helper ++, og et tilhørende softwaresystem, der gør robotprogrammering intuitiv for produktionspersonale uden robotekspertise. Arbejdet er udført i samarbejde med EU-projektet TAPAS med målet om at skabe en fleksibel industriassistent. Little Helper ++ er opbygget af en Neobotix-platform, en elektrisk griber og en KUKA Light Weight Robot, og styres af Little Helper System – et lagdelt, opgaveorienteret softwaresystem med enhedernes primitive funktioner, færdigheder (skills) og hele opgaver som tre niveauer. Systemet kan betjenes via både terminal (TUI) og en iPad-baseret grafisk brugerflade (GUI), der kombineres med fysisk indlæring, hvor operatøren guider robotten til at lære koordinater. Der er udviklet et bibliotek af færdigheder, herunder klassiske robotmakroer som pick og peg-in-hole, i alt 10 robotfærdigheder, der kan sammensættes til at løse komplekse opgaver. Tilgangen er afprøvet af en testperson uden robotbaggrund, som efter en kort introduktion kunne programmere og gennemføre en industriel opgave replikeret fra Grundfos. Systemet er yderligere verificeret ved programmering og udførelse af to komplekse montageopgaver fra Grundfos, hvilket peger på, at den færdighedsbaserede, opgaveorienterede tilgang effektivt understøtter intuitiv programmering og fleksibel automatisering.
[This apstract has been generated with the help of AI directly from the project full text]
Keywords
TAPAS ; Little Helper ; KUKA LWR ; Mobile Robot ; Automation ; Skills