Change search
ReferencesLink to record
Permanent link

Direct link
Autonomous Vehicle with Obstacle Identification and Detection
Mälardalen University, School of Innovation, Design and Engineering, Embedded Systems.
2015 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

In order to allow hospital staff to spend a bigger part of their time with actual patients, this thesishas focused on how to implement an autonomous robot to do some of their work. The robot ismade to navigate in a hospital environment while pulling a hospital bed as well as avoiding obstaclesand humans. A navigation algorithm has been implemented together with a human detection. Theprototype is able to navigate in the test environment, an office landscape, in a direction decided bythe user whilst avoiding different obstacles. Since a low cost would make it easier to implementthe solution in an actual hospital environment, a low cost has been an aim of this work. Thesolution this thesis presents is a multi-sensor system which is able to navigate through an unknownenvironment without hitting any obstacles whilst maintaining a certain heading. Analog distancesensors, encoders for motor position and a camera for human detection have been implemented.This prototype also contains a first attempt at a human detection software, which could not befinished during the time of the thesis work. The results show that both sensor fusion for navigationand the navigational algorithm are fast enough to work on this system. However, at this point, thecomplete software system with the human detection is not able to run on an embedded system at asatisfactory speed.

Abstract [sv]

For att gora det mojligt for sjukhuspersonal att spendera mer tid med sina patienter sa har dethar examensarbetet fokuserat pa hur en autonom robot skulle kunna utfora en del av deras arbete.Roboten ar byggd for att navigera i en sjukhusmiljo samtidigt som den drar en sjukhussang ochundviker hinder samt manniskor. En navigationsalgoritm har implementerats tillsammans meden manniskodetektering. Prototypen klarar av att navigera i testmiljon, ett kontorslandskap, i enriktning som anvandaren har bestamt samtidigt som den undviker hinder. Under arbetet sa harmalet varit en lag slutkostnad, eftersom en lag kostnad skulle gora det mojligt att implementerasystemet pa ett riktigt sjukhus. Losningen det har arbetet presenterar ar ett multi-sensorsystem somklarar av att navigera genom en okand miljo utan att krocka med nagra hinder samtidigt som denbehaller sin riktning. Analoga avstandssensorer, motorenkodrar for positionering av motorer samten kamera for manniskodetektering har implementerats. Prototypen innehaller ett forsta forsok tillen mjukvara for manniskodetektering, vilken inte slutfordes under projekttiden. Resultaten visar attbara anvanda navigationen och hanteringen av sensordata till denna fungerar i dagslaget. Men allmjukvara kan i dagslaget inte koras pa det inbyggda systemet eftersom det blir for langsamt.

Place, publisher, year, edition, pages
2015. , 38 p.
Keyword [en]
Autonomous navigation, computer vision, human detection
National Category
URN: urn:nbn:se:mdh:diva-28302OAI: diva2:821265
External cooperation
ÅF Technology
Subject / course
2015-06-05, Gamma, Mälardalens Högskola, Högskoleplan 1, Västerås, 14:05 (English)
Available from: 2015-06-30 Created: 2015-06-15 Last updated: 2015-06-30Bibliographically approved

Open Access in DiVA

AutonomousVehicleObstacleIdentificationDetection(5565 kB)450 downloads
File information
File name FULLTEXT01.pdfFile size 5565 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Kovala, Weronica
By organisation
Embedded Systems

Search outside of DiVA

GoogleGoogle Scholar
Total: 450 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 449 hits
ReferencesLink to record
Permanent link

Direct link