Design and development of sit-to-stand trajectory and control of humanoid robot

Mohd Bazli , Bahar (2014) Design and development of sit-to-stand trajectory and control of humanoid robot. Masters thesis, UTeM.

[img] Text

Download (1MB)


Sitting position is an important feature in a humanoid robotic system as it is more stable when compared to standing position, resulting in less energy consumption since no actuator is needed to stabilize the robot. Sitting is crucial especially for humanoid robot in security and domestic robotics field where the robots are used for a long period. In order to return to standing position, sit to stand (STS) motion is needed. One of the main challenges in STS is during the lift-off; i.e. the moment when the robot’s thigh is lifted from the chair’s surface. During lift-off, a sudden change of the position of centre of mass (CoM) causes instability to the STS motion. Furthermore, the limitation of body and joint will exacerbate the problem by limiting the ability to move the CoM to appropriate position. Due to this issue, the first objective of this research is to develop and validate a system that autonomously able to identify a trajectory to transfer the CoM to an appropriate position before lift-off from any chair height. The method works by autonomously calculate the horizontal distance between the CoM and the support polygon. With the estimated distance, flexion of hip and ankle joints is made to bring the CoM into the support polygon. The arrangement of the motion is based on Alexander STS technique. Second objective is to develop and validate a control system to balance the robot from tumbling down during STS motion due to stability issue. The proposed control system employs the IF-THEN rules as the action selector. The rules are set based on CoP position and feedback from body’s angular direction in y-axis on sagittal plane. The rules set three variable i.e. HAT (Head-Arm-Torso) direction, HAT velocity, and proportional controller gain. To determine the gain for the proportional controller, the gain identification method implements the partitioning of CoP position into a number of regions. The coefficient at each region is set differently to increase the sensitivity of the controller. To verify the effectiveness of the proposed method, experiments using NAO robot were conducted. The stability of the robot was measured based on the position of Centre of pressure (CoP) within the feet area and the angle y reading. Results show that the robot was able to perform the STS motion when height of chair is varied from 9.95cm to 16.25cm. The CoP position also shows that the pressure point is always within the feet area. However, the system failed to perform the task when the height of chair is lower than 96.60% of the robot’s shank length due to the robot’s body limitation.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Robotics, Robots
Subjects: T Technology > TJ Mechanical engineering and machinery
Divisions: Library > Tesis > FKE
Depositing User: Norziyana Hanipah
Date Deposited: 29 Jul 2015 06:38
Last Modified: 29 Jul 2015 06:38
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item