Iraj Mantegh

A PbD approach for learning pseudo-periodic robot trajectories over curved surfaces

This paper provides a new method for modeling, clustering, and generalizing complex pseudo-periodic motions in a Robot Programming by Demonstration (PbD) framework. Relevant features of the trajectories are extracted by applying a linear mapping off the surface part using Moving Window Principal Component Analysis. A Hidden Markov Model is used for segmentation and temporal clustering of feature data from multiple trajectories. The generalized trajectory is derived by spline fitting through the time-aligned features, followed by a reverse mapping back onto the working surface. The proposed approach is tested with an experiment in a highly manual manufacturing process (shot peen forming), and it is shown that the generalized paths are both more consistent and more effective than those observed in the human demonstrations.

TRAJECTORY LEARNING BASED ON CONDITIONAL RANDOM FIELDS FOR ROBOT PROGRAMMING BY DEMONSTRATION

This work presents an approach for implementation of conditional random fields (CRF) in transferring motor skills to robots. As a discriminative probabilistic model, CRF models directly the conditional probability distribution over label sequences for given observation sequences. Hereby, CRF was employed for segmentation and labeling of a set of demonstrated trajectories observed by a tracking sensor. The key points obtained by CRF segmentation of the demonstrations were used for generating a generalized trajectory for the task reproduction. The approach was evaluated by simulations of two industrial manufacturing applications.

A data fusion approach for multi-camera based visual servoing

Accurate visual servoing depends extensively on the quality of pose estimation. Sensor fusion provides a solution to improve accuracy and robustness of pose estimation. This paper introduces sensor fusion methods using two cameras to reduce the inaccuracy of pose estimation. Simulation results are reported to verify the efficiency of the proposed methods.

Path Planning for Autonomous Mobile Robots Using the Boundary Integral Equation Method

Path planning is a fundamental problem in mobile robotics. The objective of path planning is to find a sequence of states that carries a robot from some start state to the goal while avoiding obstacles. This paper focuses on point-to-point path planning, ie., developing a path between given points in a known environment with obstacles. A novel environment representation method is introduced which utilizes the Boundary Integral Equation (BIE) method to map the environment into continuous Harmonic potential fields. Different mappings and mechanisms for utilizing these mappings for path planning are developed, and a collection of path planning algorithms are described which exploit the properties of Harmonic fields and the computational efficiency of the BIE method.

Intelligent Thermal Control of Resistance Welding of Fiberglass Laminates for Automated Manufacturing

A novel approach to automate large-scale resistance welding of thermoplastic composite materials, based on real-time temperature control, is presented in this paper. Resistance welding, a kind of fusion bonding, is a widespread alternative process to mechanical fastening and adhesive bonding in joining thermoplastic composites. A second-order linear observer was identified to estimate the online temperature at the weld interface. A fuzzy logic controller (FLC) was designed to control the weld temperature using both process voltage and linear weld speed as the control inputs and using the temperature estimate as the feedback signal. The FLC was implemented in real time with the experimental setup using an industrial robot-KUKA KR 210. The closed-loop controller was capable of successfully controlling the weld temperature within 10% of the desired temperature, resulting in superior weld quality when compared to open-loop results. The results thus affirm the suitability of the control strategy for large-scale automated welding.

Knowledge-Based Task Planning Using Natural Language Processing for Robotic Manufacturing

Robotic alternative to many manual operations falls short in application due to the difficulties in capturing the manual skill of an expert operator. One of the main problems to be solved if robots are to become flexible enough for various manufacturing needs is that of end-user programming. An end-user with little or no technical expertise in robotics area needs to be able to efficiently communicate its manufacturing task to the robot. This paper proposes a new method for robot task planning using some concepts of Artificial Intelligence. Our method is based on a hierarchical knowledge representation and propositional logic, which allows an expert user to incrementally integrate process and geometric parameters with the robot commands. The objective is to provide an intelligent and programmable agent such as a robot with a knowledge base about the attributes of human behaviors in order to facilitate the commanding process. The focus of this work is on robot programming for manufacturing applications. Industrial manipulators work with low level programming languages. This work presents a new method based on Natural Language Processing (NLP) that allows a user to generate robot programs using natural language lexicon and task information. This will enable a manufacturing operator (for example for painting) who may be unfamiliar with robot programming to easily employ the agent for the manufacturing tasks.© 2010 ASME

3D Path Planning and Real-Time Simulation for Robot Manipulators With Applications to Aerospace Manufacturing

This paper presents a practical scheme for path and trajectory generation with applications in real-time simulation of robotic systems. The path planning is solved as a series of local obstacle avoidance problems which produce via points in the 3D Cartesian space as well as the normal vector to the obstacle surface at each via point. This vector is later utilized for trajectory generation. An intermediate orientation representation is used to guarantee continuity at the angular acceleration level for real-time simulation and control purposes. The proposed method is implemented in a real-time simulation environment for testing and final application. The real-time data generation allows hardware-in-the loop simulation, which is crucial in many aerospace applications.Copyright

Robotic Learning of Manipulation Tasks from Visual Perception Using a Kinect Sensor

—The paper addresses the problem of transferring new skills to robots from observation of human demonstrated skill examples. An approach is presented for retrieving trajectories of an object, being manipulated during the demonstrations, from Kinect-provided measurements. The problem of object tracking across the image frames is solved by using weighted dynamic template matching with normalized cross-correlation. Such approach takes advantage of the simultaneous image and depth measurements by the Kinect device in leveraging the pattern localization and pose estimation. Demonstrated trajectories are stochastically encoded with hidden Markov model, and the obtained model is exploited for generation of a generalized trajectory for task reproduction. The developed methodology is experimentally validated in a real-world task learning scenario.

Thermal Control of Continuous Resistance Welding of Fiberglass Laminates for Automated Manufacturing

Resistance welding is an alternative process to mechanical fastening and adhesive bonding in joining thermoplastic composites. The resistance welding procedure involves the Joule heating of a heating element located between two adjoining thermoplastics. The application of heat causes the material in the interface between two surfaces to melt. The current stops then and the interfacing surfaces bound under a compaction pressure, while cooling. This results in a fusion bond between the two components. A fuzzy logic controller is designed and implemented in real-time using the temperature estimate as the feedback signal. The result is a controller capable of maintaining the weld temperature within 10% of the reference temperature.Copyright

3D Predictive Model for Controlled Resistance Welding of Composites for Manufacturing Applications

A three-dimensional transient heat transfer model is developed for a sequential joining process (resistance welding) applied on thermoplastic composites. This process involves with moving a voltage source along a heating element that conducts the power throughout a resistive mesh, generating heat and melts and bounds two composite surfaces. The model developed here is used to predict the spatial and temporal variations in the current and temperature over the weld seam for different set of input variables. The model integrates both the resistive and thermal behaviours of components involved. The significance of this modeling approach is that it captures the movement of the electrical connection, simulating a sequential joining process along a continuous weld seam. The modeling results are compared with experimental data obtained by thermocouples and infrared camera, and accurately predict the trend of variations in weld temperature.© 2009 ASME