Sussex University Bumble Bee & Ant Simulator
This system allows simulations of bumble bee flight paths and ant tragectories to be performed. The system was designed for the University of Sussex CCNR (Centre for Computational Neuroscience and Robotics) who are conducting ongoing research into the behaviour of bumble bee and ant brains. In particular the research is focusing on the visual navigation of these animals through large scale environments. Neurological algorithms representing a Bees' brain are loaded into the system and over time evolved into more complex control systems.
A number of cameras can be mounted on the end of the Z axis to allow visual feedback from the 'simulated bee/ant' to be analysed and a decision made based on the available visual data. The robot is a large XYZ gantry system (4000 * 3500 * 2500mm) with an open workspace allowing the camera to be moved freely throughout the system. Images from the cameras are captured by the control PC, processed and analysed in real-time to determine the next decision to make and where to move the robot axes.
OpenGL (3D computer graphics) simulations of the robot are performed in real-time as the robot moves about the workspace to detect possible collisions during the dynamic vision based moves.
Dennard 2055 Pan & Tilt camera
The Dennard camera is a full 360 degree, continuous rotation pan/tilt camera that allows the robot to look in any direction within the enclosure.
Neuronics VCAM 360 degree camera
The VCAM 360 camera gives a full 360 degree view around the camera. This circular image is then 'unwrapped' into one long panaramic view of the surrounding area.
Omnitech Fisheye camera
The Omnitech fisheye lens gives an image closer to that of an insects and can provide more accurate insect simulation patterns.
Main Features
- Large open 'walk-in' workspace 4000 * 3500 * 2000mm
- Servo-motor driven 3m/sec axes with an error of less than +/- 0.1mm
- Interchangable camera systems
- Safety lockout switches for safe entry into the system
- Up to 100 move corrections per second
Research Papers
This system is under constant development by the University of Sussex. More information can be found here. Papers include:Smith, L., Philippides, A., Graham, P., Baddeley, B. and Husbands, P. (2007). Linked local navigation for visual route guidance. Biologically Inspired Robotics: A Special Issue of Adaptive Behavior, 15(3), p257-271. [abstract]
Smith, L., Philippides, A. and Husbands P. (2006). Navigation in Large-Scale Environments Using an Augmented Model of Visual Homing. In S. Nolfi et al. (Eds.), From Animals to Animats 9: Proceedings of the 9th International Conference on Simulation of Adaptive Behaviour (SAB06), 361, (pp. 251-262). Springer Berlin/Heidelberg. [abstract]
Quinn, M., Smith, L., Mayley, G. and Husbands P. (2003). Evolving controllers for a homogeneous system of physical robots: Structured cooperation with minimal sensors. Philosophical Transactions of the Royal Society of London, Series A: Mathematical, Physical and Engineering Sciences, 361, 2321-2344. [abstract]
Smith, L. and Husbands, P. (2002). Visual landmark navigation through large-scale environments. In EPSRC/BBSRC International Workshop on Biologically-Inspired Robotics: The Legacy of W. Grey Walter, 272-279. [abstract]
Please note: no bees where harmed in the making of this system
