Use this URL to cite or link to this record in EThOS:
Title: Fusion of low-cost and light-weight sensor system for mobile flexible manipulator
Author: Chang, Tai Kiang
ISNI:       0000 0004 7959 8737
Awarding Body: University of Nottingham
Current Institution: University of Nottingham
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
There is a need for non-industrial robots such as in homecare and eldercare. Light-weight mobile robots preferred as compared to conventional fixed based robots as the former is safe, portable, convenient and economical to implement. Sensor system for light-weight mobile flexible manipulator is studied in this research. A mobile flexible link manipulator (MFLM) contributes to high amount of vibrations at the tip, giving rise to inaccurate position estimations. In a control system, there inevitably exists a lag between the sensor feedback and the controller. Consequently, it contributed to instable control of the MFLM. Hence, there it is a need to predict the tip trajectory of the MFLM. Fusion of low cost sensors is studied to enhance prediction accuracy at the MFLM's tip. A digital camera and an accelerometer are used predict tip of the MFLM. The main disadvantage of camera is the delayed feedback due to the slow data rate and long processing time, while accelerometer composes cumulative errors. Wheel encoder and webcam are used for position estimation of the mobile platform. The strengths and limitations of each sensor were compared. To solve the above problem, model based predictive sensor systems have been investigated for used on the mobile flexible link manipulator using the selected sensors. Mathematical models were being developed for modeling the reaction of the mobile platform and flexible manipulator when subjected to a series of input voltages and loads. The model-based Kalman filter fusion prediction algorithm was developed, which gave reasonability good predictions of the vibrations of the tip of flexible manipulator on the mobile platform. To facilitate evaluation of the novel predictive system, a mobile platform was fabricated, where the flexible manipulator and the sensors are mounted onto the platform. Straight path motions were performed for the experimental tests. The results showed that predictive algorithm with modelled input to the Extended Kalman filter have best prediction to the tip vibration of the MFLM.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: TJ Mechanical engineering and machinery