Crane collision modelling using a neural network approach

Abstract

Abstract--The objective of the present work is to find a Collision Detection algorithm to be used in the Virtual Reality crane simulator (UVSim ®), developed by the Robotics Institute of the University of Valencia for the Port of Valencia. The method is applicable to boxshaped objects and is based on the relationship between the colliding object positions and their impact points. The tool chosen to solve the problem is a neural network, the Multilayer Perceptron (MLP), which adapts to the characteristics of the problem, namely, nonlinearity, a large amount of data, and no a priori knowledge. The results achieved by the neural network are very satisfactory for the case of box-shaped objects. Furthermore, the computational burden is independent from the object positions and how the surfaces are modelled; hence, it is suitable for the real-time requirements of the application and outperforms the computational burden of other classical methods. The model proposed is currently being used and validated in the UVSim Gantry Crane simulator. Keywords: Neural networks, real-time, simulation, collision detection INTRODUCTIONThe evolution of the hardware and techniques in the field of computer graphics has led to a widespread use of simulators and real-time environments for several purposes. One of the most rapidly growing fields of application is Virtual Reality and driving simulation systems. Such simulation systems are composed of modules that perform different tasks in order to be able to provide the user of the equipment with a feeling of total immersion. The main parts of a simulation system are the visual model, the dynamic model, and the audio-visual and inertia devices, which provide the user with different stimuli.By means of numerical, mathematical algorithms, the dynamic model calculates the position of the different bodies of the scenario. The visual model has a database that includes all the objects and visual information of the scenario. It uses the state of the simulated system computed by the dynamic model to generate the visual representation of the scene, which is projected by the projection system. To achieve a high degree of realism, an inertia simulation system composed of a mobile platform uses the accelerations computed by the dynamic system to give the user the feeling of movement.Within this field, the Robotics Institute of the University of Valencia has developed UVSim ® (http://robotica.uv.es/grupos/artec/English/simu.html). UVSim comprises several crane simulators that are currently being used for the training of stevedores working in the Port of Valencia [3]. This is the second most important harbour of Spain in terms of traffic. The simulated machinery, the rubber tyre gantry crane and the quayside crane, are used for the moving and stacking of containers in a harbour terminal, as well as for loading and unloading containers from a container ship. Figure 1 shows some images obtained during a simulation session.

Using Artificial Neural Network to Provide Realistic Lifting Capacity in the Mobile Crane Simulation

“…There were four studies that used ANNs for generating realistic feedback based on what the user does. Specifically for VR port simulations, García-Fernández et al [8] used the ANN to generate realistic impact of two colliding virtual containers based on their positions and the point of collision. They trained the ANN using collision models on different points of collision.…”

Multi-agent based traffic simulation and integrated control of freeway corridors: Part 2 integrated control optimization

“…In detail, the design point is extracted by DOE and the responses of traffic volume and delayed vehicle density for each design point are obtained by MATDYMO. Data from MATDYMO are adapted to approximate the objective function and constraints by the theory of neural network [4] and are used to normalize the optimization design. This optimization design is realized by using a simulated annealing algorithm [5] for the approximated formulation.…”