1,721,130 research outputs found
Development of a human-robot cooperative control system
No full textDuring the final year of study in the course for the Bachelor’s in Electrical & Electronics Engineering in Nanyang Technological University, the student was assigned to the Final Year Project for Safe Human-Robot Interaction under the supervision of Associate Professor Cheah Chien Chern.
The project’s objective was achieved through the use of a Microsoft Kinect depth sensor to monitor the position of a SCARA robot and determine safety actions to be taken by the robot. The Kinect was programmed through the Processing IDE on the student’s laptop while the SCARA robot was pre-programmed in Visual C++ on the robotics lab’s Windows 2000 PC. Communication between the two computers was achieved through the use of a router to establish a private network between the computers.Bachelor of Engineerin
Development of an ankle exoskeleton system
No full textIn the recent decades, there have been more research and development (R & D) on exoskeletons. An exoskeleton is an external robotic suit, donned by the user, which can either increase the user’s speed or track the user’s motion while carrying an external load. Exoskeleton R & D has been gaining popularity due to its wide range of application and improvements in enabling technology (for e.g. improved energy density of batteries). One of the applications is in the military, where the exoskeleton is used primarily to enhance the speed of the soldier or improve their endurance by helping to support the load of the soldier. This project was initiated by the author himself and was done with the direct supervision of his supervisor, Assoc Prof. Cheah Chien Chern. The objective of the project is to develop an ankle exoskeleton which tracks the user’s ankle motion via sensing of the user’s intention through the measurement of the user’s heel contact force. This report discusses the process of developing the ankle exoskeleton from scratch, from the design of hardware to the testing and implementation of the controller. Key phases of the project are hardware phase, software and electronics phase and the controller design phase. The final ankle exoskeleton was able to track the user’s ankle motion while the user oscillates his ankle, up to a maximum speed of 0.714Hz without impeding the user’s motion.Bachelor of Engineerin
Energy saving control of robot
No full textRobot, a computer-controlled machine that is programmed to move, manipulates objects, and accomplishes work while interacting with its environment. The inspiration for the design of a robot manipulator is the human arm, but with some differences. Robots are able to perform repetitive tasks more quickly, cost efficiency, accurately than human forces and energy saving[1]. Energy saving as known as energy efficiency[2] is very important for every mechanical or electrical device such as robot. In this research, Region reaching/tracking controller[3-5] and Slotine & Li controller[6] are implements into SONY SCARA[7] separately. SCARA Robot energetic consumption comparison will to make in this research to figure out which controller will consume least energy in doing a similar tasks. Research of robotics aims to realize some aspects of human functions into a mechanical system. It is interesting to observe in our daily life activities that the desired target for our reaching movements is a region rather than a point or trajectory. For example, control of robot to put an object into a cup, throwing a dart to a dart board, picking up an object from a conveying belt, driving a car along a road etc. The region reaching controller are focusing on reaching on the stationary region, for example, put a straw in a cup. Author will show that region reaching controller is a suitable controller for energy efficiency purposes compare to Slotine & Li controller. However, in some applications such as driving a car along the road, shooting a moving targer etc, the desired region is not a static region but a moving region. So, region tracking is rather more practically used in daily life. Hence, later part of the research, author will implement the region tracking controller in the moving region, measure the energy used and compare it with the Slotine & Li controller.Bachelor of Engineerin
Novel low level local features for 3D expression invariant face recognition
No full textIn this paper, we present a system based on novel low level local features to recognize 3D faces under varying facial expressions. Our local features are obtained by combinatorially selecting two points from expression insensitive semi-rigid portions of the face. The curve length between the two points is computed and the distribution of such curve lengths is used as a feature vector to model the geometric shape distribution of the face. Our proposed features are very simple to compute yet highly distinctive and discriminating. Kernel Fisher discriminant analysis is used for feature optimization, followed by a linear support vector machine classifier for recognition. The system is extensively tested on 2500 facial scans of BU 3DFE dataset. Our experimental results show that the proposed system achieves a very high average classification rate of 99.17% and verification rates of 99.0% and above for a false acceptance rate of 0.001
Development of a calibration-free vision based robot control system
No full textThe author worked on using calibration-free visual feedback from camera sensors that were not calibrated and based on this visual feedback data obtained, a neural network was developed to approximate the Jacobian matrix. This was then implemented into a standard PD controller to control a robotic arm to perform setpoint task.
Not having to calibrate the kinematics of a robot or of their sensors allows for lower maintenance cost and greater reliability. Otherwise, any changes in the camera positions or the robotic arm kinematic parameters will bring about the need for recalculation of perspective projection parameters which will be an expensive task. Aging of the robotic arm or any other maintenance work is tolerated without compromising on the robotic arm performance.
The approach used is mounting of an uncalibrated camera above the robotic arm which will observe the position of the end effector and then the respective desired position coordinates(x’, y’) will be collected and saved to be used for the offline training of the neural network. The neural network algorithm developed approximates the Jacobian matrix which is then integrated into the robot system to reach a desired setpoint. The end effector of the robotic arm will gradually move to the desired setpoint specified by a user by using the approximated Jacobian matrix.Bachelor of Engineerin
Energy-saving control of multi-robot systems
No full textMulti-robot systems are getting more popular in the area of robotics. Multi-robot systems are more likely to accomplish complicated tasks while providing efficiency and reliability compared to a single robot. Conserving energy of the robots is also important so that the robots could accomplish more tasks with the constraint of the battery capacity. Research of robotics allow one to realise that some human functions can be applied to a robot. An example would be a person walking along a path. It is an involuntary reaction that a person aim to walk anywhere along the path as long as one stay on the path. It could be seen that the goal position is actually a region instead of a specific point. In addition, one would use his eyes to look for any incoming object to change his path direction to avoid collision. These could all be applied on the robots where the robots would run in a region and using sensors to avoid collision when running in a group. The robots are set to run with conventional trajectory and region trajectory. Conventional trajectory is a traditional method for the robot to the goal position by going to a specific point. Region trajectory control can be visualised as robot travelling within a defined playing field. The robot can travel anywhere within the defined region. The author also set different robot formation to run within the designated region. In order to for multi-robots to run in an orderly manner, a certain pattern need to be considered. In this project, it is required to find out if the robot formation is a factor for energy consumption. This project focus on power consumption of the robots. As the robots are battery-powered, energy measurements would be taken using a power analyser while running the trajectories. Energy consumption will be calculated with the current and voltage values from the power analyser. Energy comparison will be make for different robot formations for different trajectories. There are some constraints faced due to the nature of the robot and the software thus it is necessary to maximize the utility of the robot to meet the goals and also to save on power consumption.In conclusion, a multi-robot system is particularly useful for completing tasks without much supervision. More power-saving controllers could be researched and developed to allow robots to run for a longer time.Bachelor of Engineerin
Vision-based control of robot
No full textToday, the use of robotic systems has significantly contributed to the increase in speed and precision of automated tasks in industrial manufacturing process. Due to the advancement of technology, it becomes highly desirable to make robots intelligent so that they can function independently with minimal inputs from users.
There are two main forms of feedback used in robot application. One is visual feedback and the other one is Cartesian feedback. Visual feedback is more robust than Cartesian feedback. However, practically, the robot working space cannot be covered totally by camera. Thus the combination of visual feedback and Cartesian feedback becomes a popular way to drive the robot.
This project serves to explore the application that robot control motion can be applied more efficiently by integration both visual feedback and Cartesian feedback. The region, in which the visual feedback is used, is specially defined. It can be considered as the camera field, and the region is fixed and not varied as the various destination points.
As part of the integration, this report also explores the application that the robot can reach the destination by passing through one occlusion area. The visual feedback cannot be applied within the occlusion area.Bachelor of Engineerin
Robotic formation control In micro world
No full textOptical Tweezers can be used not only for the positioning of cells but also for moving them over centimeter distances by using laser beams of defined alignment. Micro-manipulation means controlling microscopic test objects using micro-sized equipment such as optical tweezers.
The aim of this paper is to discuss the key concepts in deriving the different controllers and to design a semi-automated controller that enables both machine and human to take charge, as well as avoid obstacles when needed, through human-machine interaction by using optical tweezer system.
The results obtained show that human-machine interaction is a combination of both automatic and human-guided approach which can be ground-breaking to micro world.
While there are limitations which is related to computation time and accuracy, the integration of different controllers is an interesting and helpful approach in helping us to explore and understand the meaning of co-manipulation.Bachelor of Engineerin
Region reaching control of robot manipulators
No full textIn this thesis, a new control con- cept called region reaching control for robot manipulators is proposed. In this new control concept, the desired objective can be specified as a region instead of a point. Several region reaching controllers are proposed in both joint-space and task-space. Since the desired region can be specified arbi- trarily small, the region reaching control concept is also a generalization of setpoint control problem.MASTER OF ENGINEERING (EEE
A unified control framework for human-robot interaction
No full textCo-existence of human and robot in the same workspace requires the robot to perform robot tasks such as trajectory tracking and also interaction tasks such as keeping a safe distance from human. According to various human-robot interaction scenarios, different interaction tasks usually require different task requirements or specifications, leading to different control strategies. Besides, due to different natures of the robot tasks and interaction tasks, different controllers may be required when the task is switched from one to another. So far, there is no theoretical framework which integrates different robot and interaction task requirements into a unified robot control strategy.
In this research, a general human-robot interaction control framework is proposed for the scenario of human and robot coexisting in the same workspace. We propose a general potential energy function which can be used to derive a stable and unified controller for various robot tasks and human-robot interaction tasks. Instead of designing a particular task function formalism for each subtask requirement, various tasks can be specified at a user level through simply adjusting certain task parameters. Interactive weights are also defined to specify the interaction behaviours of robots according to different human-robot interaction applications. Specific interaction modes such as human-dominant interaction and robot-dominant interaction are given in details to demonstrate the applications of the proposed control method. We show how the control framework can be applied to existing robot control systems with velocity control or torque control mode by developing a joint velocity reference command and an adaptive controller.
Typically, industrial manipulators have closed architecture control systems and do not come with external sensors. During human-robot interaction, robots are operated in an uncertain environment with presence of humans and are required to adjust the behaviours according to human's intentions. Hence, external sensors such as vision systems must be added and integrated into the robots to improve their capabilities in perception and reaction. Since different configurations and types of sensors result in different sensory transformation or Jacobian matrices and thus lead to different models, it is in general difficult for operators or users in factory to model the sensory systems and deploy the robots according to various human-robot interaction applications. In this thesis, a new learning algorithm is derived and employed in the proposed control framework to estimate the unknown kinematics such that various external sensors can be easily integrated into the proposed framework to perform interaction tasks without modeling the kinematics.
In the proposed framework, the robot's behaviours during the interaction can be varied by manually adjusting the task parameters. As some of the task parameters do not correspond to any physical meaning, it may be difficult for normal non-expert users to set the task parameters according to a specific interaction task. On the other hand, it is anticipated that task specification through human's demonstrations would be one of the effective ways for robots to understand or imitate human's behaviours, especially during human-robot interactions. Therefore, a task requirement learning algorithm is proposed where the motion behaviours demonstrated by human can be acquired or learned by the robot systems in a unified way.Doctor of Philosoph
- …
