A 3D simulator for autonomous robotic fish

Abstract

This paper presents a 3D simulator used for studying the motion control and autonomous navigation of robotic fish. The simulator's system structure and computation flow are presented. Simplified kinematics and hydrodynamics models for a virtual robotic fish are proposed. Many other object models are created for water, obstacles, sonar sensors and a swimming pool. Experimental results show that the simulator provides a realistic and convenient way to develop autonomous navigation algorithms for robotic fish.

The skin deformation of a 3D virtual human

“…From (2), the weights satisfy the relation: when d = d1, ω 1 1 = 1, ω 2 1 = 0; while when d = 0, ω 1 1 = ω 2 1 = 0.5 is found. The nearest the point p approaches the knee joint point p0, the less influence ω 1 1 produces and the more influence ω 2 1 produces.…”

“…Surface models of 3D virtual humans consist of polygonal meshes, which are also called model skin [1,2] . In most application cases, the surface meshes can generate skin deformation.…”

Implementation of a Localization-Oriented Hri for Walking Robots in the Robocup Environment

“…As in previously simulated applications [Liu and Hu, 2004], a three-layered object definition architecture is implemented for the localization process and user software interface. In this case, each layer has a specific purpose in architectural design, as shown in Fig.…”

Biologically inspired behaviour design for autonomous robotic fish

“…To prove the feasibility of follow-wall behaviour, it is tested in a simulator proposed in [24]. All parameters of follow-wall behaviour are given as same as in this section.…”