Presented in this paper is a method that combines computer and human intelligence into a joint power to confront the intractability of assembly sequence planning. This method applies the theory of parameterized complexity to identify the source of complexity, and then separates and designates different tasks according to the strengths of autonomous planning and human reasoning for collaborative planning. The injection of human expertise via collaboration greatly helps to reduce the computational complexity of autonomous planning, and parameterized complexity analysis provides the much needed guideline for automated processes to decide when and what types of help are needed so that human intervention can be reduced to the minimum.

The current market of computer products is large in quantity, competitive in quality, and fast in changes. This paper develops a technique to enable early evaluation of product performance on computer hardware and/or software systems with selectable factors in respect of certain aspects of product quality. By applying statistical analysis, this new technique dramatically reduces the need of experiments that are deemed necessary for product evaluation. The speed up in processing time makes this technique particularly useful in online shopping to assist consumers making purchasing decisions and in software development to help software engineers achieving optimal designs.

This paper develops a new approach of interactive assembly planning. This approach provides an intuitive hand-based interface for human operators to perform assembly operations directly in the virtual world. Its interactive planner then extracts the knowledge of mechanical assembly, and determines motion paths at run time with a biologically inspired neural network. From a single user-defined assembly sequence, this approach is able to automatically produce alternative assembly sequences with robot-level instructions.

In this paper, a neural network based torque controller is proposed for real-time collision-free navigation of nonholonomic mobile robots. A torque resulted from the obstacles is incorporated in the control design based on the artificial potential technique, which locally pushes the robot away from the obstacles to avoid collisions. All the needed environment information can be obtained from on-board robot sensors that have limited visibility range only. A torque from a simply single-layer neural network is employed to learn the completely unknown robot dynamics. The system stability is guaranteed by a Lyapunov stability theory. The real-time fine control of mobile robots is achieved through the on-line learning of the neural network. The effectiveness of the proposed controller is demonstrated by simulation studies in both static and dynamic environments.

This paper investigates the introduction of biologically inspired intelligence into virtual assembly. It develops a approach to assist product engineers making assembly-related manufacturing decisions without actually realizing the physical products. This approach extracts the knowledge of mechanical assembly by allowing human operators to perform assembly operations directly in the virtual environment. The incorporation of a biologically inspired neural network into an interactive assembly planner further leads to the improvement of flexible product manufacturing, i.e., automatically producing alternative assembly sequences with robot-level instructions for evaluation and optimization. Complexity analysis and simulation study demonstrate the effectiveness and efficiency of this approach.

An area-covering operation is a kind of complete coverage path planning, which requires the robot path to cover every part of the workspace, which is an essential issue in cleaning robots and many other robotic applications such as vacuum, robots, painter robots, land mine detectors, lawn mowers, and windows cleaners. In this paper, a novel biologically inspired neural network approach is proposed for complete coverage path planning with obstacle avoidance of a cleaning robot in a nonstationary environment. The dynamics of each neuron in the topologically organized neural network is characterized by a shunting equation or an additive equation derived from Hodgkin and Huxley's (1952) membrane equation. There are only local lateral connections among neurons. Thus the computational complexity linearly depends on the neural network size. The proposed model algorithm is computationally efficient, and can also deal with changing environment. Simulation results show that the proposed model is capable of planning collision-free complete coverage robot path.

An interactive approach to assembly planning is presented. It provides a virtual reality interface for production engineers to program the virtual representation of robotic manipulators in a three-dimensional (3D) operation space. The direct human involvement creates a user-defined assembly sequence, which contains the human knowledge of mechanical assembly. By extracting the precedence relationship of machinery parts, for the first time it becomes possible to generate alternative assembly sequences automatically from a single sequence for robot reprogramming. This interactive approach introduces human expertise into assembly planning, thus breaking down the computational complexity of autonomous systems. Experiments and analysis provide strong evidence to support the incontestable advantages of "manufacturing in the computer".

In this paper, shunting neural networks are proposed for dynamic planning and control of robots. The dynamic environment is represented by a neural activity landscape of a neural network, where each neuron in the topologically organized neural network is characterized by a shunting equation that is derived from Hodgkin and Huxley's (1952) biological membrane equation. The collision-free path is generated in real-time from the activity landscape without any explicit searching procedures and without any prior knowledge of the dynamic environment. The real-time tracking control of robots to follow the planned dynamic path is designed using shunting equation as well. The effectiveness and efficiency of the proposed approach are demonstrated through simulation and comparison studies. Simulation in several computer-synthesized virtual environments further demonstrates the advantages of the proposed approach with encouraging experimental results.

This paper presents a novel approach toward the real-time rendering of artificial guidance in virtual programming. It models the operational characteristics of robot programming with a biologically inspired neural network. Through neural activity propagation, this approach creates a new means of assisting the direct manipulation of virtual objects with suggested moving paths. In addition, it introduces a tracking controller to ensure smooth moving velocities. Finally, simulation results are provided to demonstrate the effectiveness and efficiency of the proposed approach.

In this paper, a novel biologically inspired neural network approach is proposed for real-time motion planning with obstacle avoidance of a nonholonomic car-like robot in a nonstationary environment. The dynamics of each neuron in the topologically organized neural network is characterized by a shunting equation derived from Hodgkin and Huxley's (1952) membrane equation. The robot configuration space constitutes the state space of the neural network. There are only local connections among neurons. Thus the computational complexity linearly depends on the neural network size. The neural activity propagation is subject to the kinematic constraints of the nonholonomic car-like robot. The real-time robot motion is planned through the dynamic neural activity landscape without any prior knowledge of the dynamic environment, without any learning procedures, and without any local collision checking procedures at each step of the robot movement. Therefore the model algorithm is computationally efficient. The stability of the neural network system is proved by qualitative analysis and a Lyapunov stability theory. Simulation in several computer-synthesized virtual environments further demonstrates the advantages of the proposed approach with encouraging experimental results.

This paper investigates assembly planning in virtual environments. It introduces a virtual teaching pendant and develops an interactive planning algorithm. This algorithm is capable of extracting human expertise of mechanical assembly from human-computer interaction and then using it for the automatic generation and interactive evaluation of assembly sequences. Analysis shows that the proposed interactive approach has a low complexity and hence offers a valid alternative to the usual autonomous approaches of assembly planning.

Virtual assembly computer tools assist human operators in their work of mechanical assembly inside virtual environments. This paper investigates the real-time rendering of computer-synthesized guidance. It uses a biologically inspired neural network to generate collision-free paths, and helps users to specify assembly tasks by guiding the motion of objects. This work leverages the power of "manufacturing in the computer" with the creation of artificial guidance that could be unavailable in the physical world.

Working in the virtual environment has the advantage of unsurpassed support from computer tools. However, its success in industrial applications still relies on how accurately a virtual environment represents the real-life activities. Presented in this paper is a method of reactive assembly planning that is able to handle environmental uncertainty. It uses automatic object modeling to update the virtual environment according to the changes in a working environment, and a biologically-inspired neural network to reactively generate collision-free paths in real time.

This paper investigates the integration of robotic task programming and assembly sequence planning in virtual assembly. It presents an integrated approach that derives assembly relationship of machinery parts as required by sequence planning automatically from the information provided by interactive programming. 'Automatic reprogramming' also becomes accessible, which feeds alternative assembly sequences back to programming for optimization.

The symbolic representation of 3D objects is the fundamental knowledge for computer systems to understand the environment. This knowledge is usually assumed to exist in a computer but can also be acquired by accumulating spatial features extracted from sensory inputs at different viewing directions. This paper first investigates surface visibility and, then, after introducing mass vector chains (MVC), discusses the relationship between MVC and the spatial closure of object models. An automatic modeling mechanism is established with the observation that the boundary of an object is closed only if the MVC of its model is closed or, alternatively, the tail-to-head vector of an unclosed MVC estimates the visible direction of the missing surfaces. Experimental results and an algorithm are also given at the end.>