17th International IEEE Conference on Intelligent Transportation Systems (ITSC), 2014
As autonomous vehicles are venturing on the urban road, reasoning other vehicles' motion inte... more As autonomous vehicles are venturing on the urban road, reasoning other vehicles' motion intention is always essential for anticipatory decision-making. This paper considers a new class of motion intention reasoning problem using cooperative perception. Specifically, thanks to the extended perception range contributed by cooperative perception, we can explicitly consider the behavior correlations among multiple vehicles, where each vehicle's motion intention can be inferred using both the temporal observations and the other vehicles' motion intention expectation. As such, a Simplified variant of Coupled Hidden Markov Model (S-CHMM) is proposed for motion intention inference. The experiment on an autonomous vehicle shows that the proposed algorithm can improve the motion intention accuracy and provide efficient overtaking triggering for autonomous driving on urban road.
One of the ways to know and to determine one's shape is through the sense of touch. Even in t... more One of the ways to know and to determine one's shape is through the sense of touch. Even in the absence of vision, it would still be possible to recognize one's shape by simply touching the surface. However, there are things that you cannot touch. And even though the use of vision may be difficult, you can still identify the object's shape or figure by using devices such as cameras. In this study, the researchers aim to feel and identify the shape of an object that you can not see, and more importantly, an object that does not physically exist. This thesis is making use of humanÂ’s haptic sense in order to feel the shape of the object without it being physically present. The shape of the objects coming from the computer will be felt by the user through a matrix of tactile pins. Each pin corresponds to one motor which is controlled by a motor driver. Data information sent to the motor driver will come from the microcontroller which is connected to the computer via RS232.
2014 IEEE International Conference on Robotics and Automation (ICRA), 2014
This paper considers the problem of motion planning for linear systems subject to Gaussian motion... more This paper considers the problem of motion planning for linear systems subject to Gaussian motion noise and proposes a risk-aware planning algorithm: CC-RRT*-D. The proposed CC-RRT*-D employs the chance-constraint approximation and leverages the asymptotically optimal property of RRT* framework to compute risk-aware and asymptotically optimal trajectories. By explicitly considering the state dependence for prior state estimate, the over-conservative problem of chance-constraint approximation can be provably solved. Computational experiment results show that CC-RRT*-D is efficient and robust compared with related algorithms. The real-time experiment on an autonomous vehicle shows that our proposed algorithm is applicable to real-time obstacle avoidance.
To get the best features of both deliberative and reactive controllers, present mobile robot cont... more To get the best features of both deliberative and reactive controllers, present mobile robot control architectures are designed to accommodate both types of controller. However, these architectures are still very rigidly structured thus deliberative modules are always assigned to the same role as a high-level planner or sequencer while low-level reactive modules are still the ones directly interacting with the
Abstract In 2005 the Faculty of Engineering at the National University of Singapore relaunched a ... more Abstract In 2005 the Faculty of Engineering at the National University of Singapore relaunched a master's of science program in the management of technology (MSc/MOT) after it was moved from the School of Business where, as the first of its kind in Asia, it was started in 1992. The opportunity to relaunch the program in the faculty of engineering provided a new opportunity to refresh the program and introduce management of technology (MOT) courses that were not feasible when the program resided in the School of Business. ...
Abstract. A mobile manipulation system often involves combining more than one robot together, typ... more Abstract. A mobile manipulation system often involves combining more than one robot together, typically a manipulator arm and a mobile base. To implement a force and motion control with dynamic compensation, a torque-controlled system is necessary. However, a torque-controlled robot is not always available. In fact, most commercially available mo-bile bases are velocity-controlled. This paper presents a method for combining a torque-controlled arm and a velocity-controlled base, while performing a force and motion task. The operational space formulation using a consistent set of integrated arm-base robot dy-namics is employed in a mobile manipulation task of polishing an aircraft canopy. The torque controlled arm compensates for the dynamics introduced by the mobile base. The added mobility of the base enables the arm to cover the entire workspace. 1
Abstract: A solution to the singularity problem is presented from the approach of the operational... more Abstract: A solution to the singularity problem is presented from the approach of the operational space formulation which involves both motion and force control. A brief summary of the Operational Space Formulation and how singularity presents a problem is explained. The inverse of the Jacobian at singular configuration is handled by removing the degenerate components of the motion, therefore manipulator is made redundant to the task. Khatib’s [3] dynamically consistent inverse was then used to invert the Jacobian and to create a null space motion to escape from the singular configuration in the case that the desired path lies in the degenerate direction. The algorithm was implemented on the PUMA 560 manipulator, and the results are presented. 1.
2021 IEEE International Conference on Robotics and Automation (ICRA)
Navigation through dynamic pedestrian environments in a socially compliant manner is still a chal... more Navigation through dynamic pedestrian environments in a socially compliant manner is still a challenging task for autonomous vehicles. Classical methods usually lead to unnatural vehicle behaviours for pedestrian navigation due to the difficulty in modeling social conventions mathematically. This paper presents an end-to-end path planning system that achieves autonomous navigation in dynamic environments through imitation learning. The proposed system is based on a fully convolutional neural network that maps the raw sensory data into a confidence map for path extraction. Additionally, a classification network is introduced to reduce the unnecessary re-plannings and ensures that the vehicle goes back to the global path when re-planning is not needed. The imitation learning based path planner is implemented on an autonomous wheelchair and tested in a new real-world dynamic pedestrian environment. Experimental results show that the proposed system is able to generate paths for different driving tasks, such as pedestrian following, static and dynamic obstacles avoidance, etc. In comparison to the state-of-the-art method, our system is superior in terms of generating human-like trajectories.
Fifth International Conference on Advanced Robotics 'Robots in Unstructured Environments
The existence of singular positions inside the robot workspace is an inherent problem for task pl... more The existence of singular positions inside the robot workspace is an inherent problem for task planning and robot control. Identifying the singularities in the robot workspace is not always an analytically tractable problem, except for manipulators with opportune placement of the joint axes. Typically, it is easier and more informative to evaluate the robot `arm' and `wrist' singularities which reflect
2011 IEEE/RSJ International Conference on Intelligent Robots and Systems
ABSTRACT This paper presents an integrated stochastic approach to tracking multiple live cells th... more ABSTRACT This paper presents an integrated stochastic approach to tracking multiple live cells that interact with the surrounding gel matrix. Cells migrate in a stochastic manner, forming a functional structure. Tracking the trajectory of each migrating cell and its interactions with the gel and other cells provide useful insights into how a vascular structure is formed as a collection of migratory cells. In micro-fluidic 3D angiogenic sprouting experiments, two types of images are obtained at discrete time steps using confocal microscopy: a) three-dimensional fluorescent images of stained cell nuclei and b) two dimensional visible light images of the gel matrix. These two sources of images provide supplementary information as the outline of the conduit or lumen formed in the matrix by the migrating cells can be seen in the images of the gel. A Bayesian filtering framework is developed which augments both the cell and conduit parameters to the same state vector, allowing mathematically consistent simultaneous observation updates from both channels. Issues encountered include the high dimensional state vector and non-Gaussian cell state updates. Results demonstrate that our method based on Rao-Blackwellized particle filtering treats these issues effectively.
2014 International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM)
This paper describes localization and navigation in an indoor environment with transparent object... more This paper describes localization and navigation in an indoor environment with transparent objects. Although LIDAR proved to be a popular choice, its inability to detect glass surface requires another sensor to ensure navigation capability will work in all kind of environment. In this paper, sonar sensor is used in complement to laser due to its capability to detect hard surfaces and low cost. The experiment shows that a better map can be achieved when these two sensors are used.
17th International IEEE Conference on Intelligent Transportation Systems (ITSC), 2014
As autonomous vehicles are venturing on the urban road, reasoning other vehicles' motion inte... more As autonomous vehicles are venturing on the urban road, reasoning other vehicles' motion intention is always essential for anticipatory decision-making. This paper considers a new class of motion intention reasoning problem using cooperative perception. Specifically, thanks to the extended perception range contributed by cooperative perception, we can explicitly consider the behavior correlations among multiple vehicles, where each vehicle's motion intention can be inferred using both the temporal observations and the other vehicles' motion intention expectation. As such, a Simplified variant of Coupled Hidden Markov Model (S-CHMM) is proposed for motion intention inference. The experiment on an autonomous vehicle shows that the proposed algorithm can improve the motion intention accuracy and provide efficient overtaking triggering for autonomous driving on urban road.
One of the ways to know and to determine one's shape is through the sense of touch. Even in t... more One of the ways to know and to determine one's shape is through the sense of touch. Even in the absence of vision, it would still be possible to recognize one's shape by simply touching the surface. However, there are things that you cannot touch. And even though the use of vision may be difficult, you can still identify the object's shape or figure by using devices such as cameras. In this study, the researchers aim to feel and identify the shape of an object that you can not see, and more importantly, an object that does not physically exist. This thesis is making use of humanÂ’s haptic sense in order to feel the shape of the object without it being physically present. The shape of the objects coming from the computer will be felt by the user through a matrix of tactile pins. Each pin corresponds to one motor which is controlled by a motor driver. Data information sent to the motor driver will come from the microcontroller which is connected to the computer via RS232.
2014 IEEE International Conference on Robotics and Automation (ICRA), 2014
This paper considers the problem of motion planning for linear systems subject to Gaussian motion... more This paper considers the problem of motion planning for linear systems subject to Gaussian motion noise and proposes a risk-aware planning algorithm: CC-RRT*-D. The proposed CC-RRT*-D employs the chance-constraint approximation and leverages the asymptotically optimal property of RRT* framework to compute risk-aware and asymptotically optimal trajectories. By explicitly considering the state dependence for prior state estimate, the over-conservative problem of chance-constraint approximation can be provably solved. Computational experiment results show that CC-RRT*-D is efficient and robust compared with related algorithms. The real-time experiment on an autonomous vehicle shows that our proposed algorithm is applicable to real-time obstacle avoidance.
To get the best features of both deliberative and reactive controllers, present mobile robot cont... more To get the best features of both deliberative and reactive controllers, present mobile robot control architectures are designed to accommodate both types of controller. However, these architectures are still very rigidly structured thus deliberative modules are always assigned to the same role as a high-level planner or sequencer while low-level reactive modules are still the ones directly interacting with the
Abstract In 2005 the Faculty of Engineering at the National University of Singapore relaunched a ... more Abstract In 2005 the Faculty of Engineering at the National University of Singapore relaunched a master's of science program in the management of technology (MSc/MOT) after it was moved from the School of Business where, as the first of its kind in Asia, it was started in 1992. The opportunity to relaunch the program in the faculty of engineering provided a new opportunity to refresh the program and introduce management of technology (MOT) courses that were not feasible when the program resided in the School of Business. ...
Abstract. A mobile manipulation system often involves combining more than one robot together, typ... more Abstract. A mobile manipulation system often involves combining more than one robot together, typically a manipulator arm and a mobile base. To implement a force and motion control with dynamic compensation, a torque-controlled system is necessary. However, a torque-controlled robot is not always available. In fact, most commercially available mo-bile bases are velocity-controlled. This paper presents a method for combining a torque-controlled arm and a velocity-controlled base, while performing a force and motion task. The operational space formulation using a consistent set of integrated arm-base robot dy-namics is employed in a mobile manipulation task of polishing an aircraft canopy. The torque controlled arm compensates for the dynamics introduced by the mobile base. The added mobility of the base enables the arm to cover the entire workspace. 1
Abstract: A solution to the singularity problem is presented from the approach of the operational... more Abstract: A solution to the singularity problem is presented from the approach of the operational space formulation which involves both motion and force control. A brief summary of the Operational Space Formulation and how singularity presents a problem is explained. The inverse of the Jacobian at singular configuration is handled by removing the degenerate components of the motion, therefore manipulator is made redundant to the task. Khatib’s [3] dynamically consistent inverse was then used to invert the Jacobian and to create a null space motion to escape from the singular configuration in the case that the desired path lies in the degenerate direction. The algorithm was implemented on the PUMA 560 manipulator, and the results are presented. 1.
2021 IEEE International Conference on Robotics and Automation (ICRA)
Navigation through dynamic pedestrian environments in a socially compliant manner is still a chal... more Navigation through dynamic pedestrian environments in a socially compliant manner is still a challenging task for autonomous vehicles. Classical methods usually lead to unnatural vehicle behaviours for pedestrian navigation due to the difficulty in modeling social conventions mathematically. This paper presents an end-to-end path planning system that achieves autonomous navigation in dynamic environments through imitation learning. The proposed system is based on a fully convolutional neural network that maps the raw sensory data into a confidence map for path extraction. Additionally, a classification network is introduced to reduce the unnecessary re-plannings and ensures that the vehicle goes back to the global path when re-planning is not needed. The imitation learning based path planner is implemented on an autonomous wheelchair and tested in a new real-world dynamic pedestrian environment. Experimental results show that the proposed system is able to generate paths for different driving tasks, such as pedestrian following, static and dynamic obstacles avoidance, etc. In comparison to the state-of-the-art method, our system is superior in terms of generating human-like trajectories.
Fifth International Conference on Advanced Robotics 'Robots in Unstructured Environments
The existence of singular positions inside the robot workspace is an inherent problem for task pl... more The existence of singular positions inside the robot workspace is an inherent problem for task planning and robot control. Identifying the singularities in the robot workspace is not always an analytically tractable problem, except for manipulators with opportune placement of the joint axes. Typically, it is easier and more informative to evaluate the robot `arm' and `wrist' singularities which reflect
2011 IEEE/RSJ International Conference on Intelligent Robots and Systems
ABSTRACT This paper presents an integrated stochastic approach to tracking multiple live cells th... more ABSTRACT This paper presents an integrated stochastic approach to tracking multiple live cells that interact with the surrounding gel matrix. Cells migrate in a stochastic manner, forming a functional structure. Tracking the trajectory of each migrating cell and its interactions with the gel and other cells provide useful insights into how a vascular structure is formed as a collection of migratory cells. In micro-fluidic 3D angiogenic sprouting experiments, two types of images are obtained at discrete time steps using confocal microscopy: a) three-dimensional fluorescent images of stained cell nuclei and b) two dimensional visible light images of the gel matrix. These two sources of images provide supplementary information as the outline of the conduit or lumen formed in the matrix by the migrating cells can be seen in the images of the gel. A Bayesian filtering framework is developed which augments both the cell and conduit parameters to the same state vector, allowing mathematically consistent simultaneous observation updates from both channels. Issues encountered include the high dimensional state vector and non-Gaussian cell state updates. Results demonstrate that our method based on Rao-Blackwellized particle filtering treats these issues effectively.
2014 International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM)
This paper describes localization and navigation in an indoor environment with transparent object... more This paper describes localization and navigation in an indoor environment with transparent objects. Although LIDAR proved to be a popular choice, its inability to detect glass surface requires another sensor to ensure navigation capability will work in all kind of environment. In this paper, sonar sensor is used in complement to laser due to its capability to detect hard surfaces and low cost. The experiment shows that a better map can be achieved when these two sensors are used.
Uploads
Papers by Marcelo Ang