Embedded system application development on raspberry Pi 3 : smart surveillance robot
Traditional SLAM (Simultaneous localization and mapping) robots made use of techniques such as the extended Kalman filter and fastSLAM. The extended Kalman filter is a recursive technique whereby estimation of the current state is based on measurement from the previous state. Firstly, the algorit...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/77565 |
| _version_ | 1826112616569765888 |
|---|---|
| author | Lai, Ken Kin Yong |
| author2 | Chong Yong Kim |
| author_facet | Chong Yong Kim Lai, Ken Kin Yong |
| author_sort | Lai, Ken Kin Yong |
| collection | NTU |
| description | Traditional SLAM (Simultaneous localization and mapping) robots made use of techniques
such as the extended Kalman filter and fastSLAM. The extended Kalman filter is a recursive
technique whereby estimation of the current state is based on measurement from the previous
state. Firstly, the algorithm predicts the state of the robot based on some sensor values.
Subsequently, the actual state of the robot is observed based on other sensor values. Actual
state of robot may be wrong due to errors in sensor. Finally, the algorithm compares the two
values and determines the best location for the state of the robot. The process is repeated for
the subsequence state of the robot [1].
The above short explanation demonstrated the limitation of implementing such algorithms
into low cost SLAM robots as the localization depends heavily on the precision of the sensor.
A precise sensor is needed for accurate estimation. This limits the possibility of implementing
SLAM into embedded systems found commonly in our daily life. The project look into the
possibility of using the raspberry pi 3 embedded system board together with simple sensors to
implement SLAM using a grid based technique. Unlike traditional techniques, the grid base
technique is less sensitive to the precision of the sensors and will be able to better determine
the position of the robot in places with fewer landmarks. Overall, the final product of the
project is able to explore an indoor environment and produce a fairly accurate map |
| first_indexed | 2024-10-01T03:09:59Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/77565 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:09:59Z |
| publishDate | 2019 |
| record_format | dspace |
| spelling | ntu-10356/775652023-07-07T18:06:02Z Embedded system application development on raspberry Pi 3 : smart surveillance robot Lai, Ken Kin Yong Chong Yong Kim School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering Traditional SLAM (Simultaneous localization and mapping) robots made use of techniques such as the extended Kalman filter and fastSLAM. The extended Kalman filter is a recursive technique whereby estimation of the current state is based on measurement from the previous state. Firstly, the algorithm predicts the state of the robot based on some sensor values. Subsequently, the actual state of the robot is observed based on other sensor values. Actual state of robot may be wrong due to errors in sensor. Finally, the algorithm compares the two values and determines the best location for the state of the robot. The process is repeated for the subsequence state of the robot [1]. The above short explanation demonstrated the limitation of implementing such algorithms into low cost SLAM robots as the localization depends heavily on the precision of the sensor. A precise sensor is needed for accurate estimation. This limits the possibility of implementing SLAM into embedded systems found commonly in our daily life. The project look into the possibility of using the raspberry pi 3 embedded system board together with simple sensors to implement SLAM using a grid based technique. Unlike traditional techniques, the grid base technique is less sensitive to the precision of the sensors and will be able to better determine the position of the robot in places with fewer landmarks. Overall, the final product of the project is able to explore an indoor environment and produce a fairly accurate map Bachelor of Engineering (Electrical and Electronic Engineering) 2019-05-31T06:37:08Z 2019-05-31T06:37:08Z 2019 Final Year Project (FYP) http://hdl.handle.net/10356/77565 en Nanyang Technological University 70 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Electrical and electronic engineering Lai, Ken Kin Yong Embedded system application development on raspberry Pi 3 : smart surveillance robot |
| title | Embedded system application development on raspberry Pi 3 : smart surveillance robot |
| title_full | Embedded system application development on raspberry Pi 3 : smart surveillance robot |
| title_fullStr | Embedded system application development on raspberry Pi 3 : smart surveillance robot |
| title_full_unstemmed | Embedded system application development on raspberry Pi 3 : smart surveillance robot |
| title_short | Embedded system application development on raspberry Pi 3 : smart surveillance robot |
| title_sort | embedded system application development on raspberry pi 3 smart surveillance robot |
| topic | DRNTU::Engineering::Electrical and electronic engineering |
| url | http://hdl.handle.net/10356/77565 |
| work_keys_str_mv | AT laikenkinyong embeddedsystemapplicationdevelopmentonraspberrypi3smartsurveillancerobot |