This post is password protected. To view it please enter your password below:

Password:

Click for more info: 
11 Kuber  / Trans-Ice / Three Tremendous Trees

Shadowed Perception # 1 / plot, aluminium, 96 x 96 c / 2007

Shadowed Perception # 2 / plot, aluminium, 96 x 96 cm / 2007

Shadowed Perception # 3, 4, 5 / plot, aluminium, 300 x 100 cm / 2007

Shown at Trondheim Kunstmuseum 2007.

TRINIGON

The Norwegian National Monument of Freedom, Narvik / 1995-2005 / polished steel (8mm hot rolled stainless steel with lux value = mirror) / h. = 18. m.

Trinigon was raised by the Norwegian government in 1995 in connection with the 50 years anniversary of the ending of World War II. The monument is a homage to all victims on both sides of the conflict, and a symbol for the importance of freedom of speech and autonomy for all people.

The monument has been rebuild three times. The two first versions became objects of poor engineering, with the second one being blown down during a winter storm in 2001 due to a inadequate build foundation. Trinigon was then raised for the third time in 2005, 10 years after the original was put up. This last version of the monument takes care of every intention entailed in the project!

Project manager for this last version: Forsvarsbygg, Harstad, Norway.
Engineering and production: BMV AS, Borkenes, Norway.
Surface quality: Outo Kumpu AB, Avesta, Sweden.

The surface of the sculpture was by dec. 2006 considered being some of the brightest 8 mm. steel plates in the world.

TRANSCUBES

2003, A – I, 30 x 30 x 30 cm, copper. Collection of Tromsø Art Museum.

Trans-Ice
kobber, stål, gress, is / copper, steel, grass, ice
170 x 60 x 60 cm, 140 x 60 x 60 cm / 2007

Four sculptures runned by cooling units making pre designated copper pipes rendering ice on their surface. The melting of the ice act as life support for plants growing beneath.

Trans<->Former #13 is a sculpture that continously redraws it´s own positions in space. New shapes are being created in an eternal process, where the movements itself takes part as the sculpting force.

In collaboration with the Department of engineering cybernetichs, and the Department of computer and information science, NTNU, we are trying to develop computer controlled joints based on artificial intelligence for this sculpture. The assembled sculpture will then be able to change its shape by automatically repositioning the joints. The sculpture parts mutual positions can also be altered by various forms of external trigging.

The sculpture can interact with the audience at sight, transforming it´s positions and shape according to the viewers amount and distance. The sculpture is powered by solar panels and is undependent of external wiring.

The project origin from the sculpture “11 Cubes” from 1998, that is shaped out of one single square pipe that every 40 cm is bent in a 90 degrees angel following a certain algoritm.

—–

Evolution of Control system(s) for a Multi Joint Snake

Authors: Gunnar Tufte and Espen Gangvik.

Genetic algorithms are developed by the Dep. of computer and information science, NTNU, by associate professor Gunnar Tufte / master theses: Karl Hatteland.

The project is based on artist Espen Gangvik’s work on a 3-dimensional interactive sculpture Trans<->Former \# 13. Trans<->Former \# 13 can change shape by interaction with it’s enviroment e.g. audience. The sculpture can be looked at as a robot snake consisting of multiple joints. The control of the joints can be done by a computer algorithm(s) that can alter the shape of the sculpture.

The robot snake consists of x# of joints that can be controlled individually. As such, the degree of freedom is very large. As to be able to find control mechanisms for the robot snake evolutionary algorithms may be a solution.

The project have interesting properties in the field of evolutionary computation and evolutionary robotics e.g. the complexity of the robot itself and the size of the search space. Interesting topics include (but are not limited to) autonomous genetic algorithms for each joint (inspired by cellular programing) exploration of individual joint properties, artificial development, dynamic changing evaluation metric (fitness). \\

From a computer science point of view Trans<->Former \# 13 can be looked at as an robot snake consisting of autonomous robots connected together to form a snake. Looking at Trans<->Former \# 13 as a collection of autonomous robots gives the opportunity to apply biological inspired computation methods from the field of cellular computation. Such machines consists of a collection of relative simple computation elements (cells) that only communicate with their neighbouring cells. The result of the computation emerges as a global behaviour given by the local action of all the cells. As such the result of the computation is the form of Trans<->Former \# 13.

At present time the computer science part of the project aims to apply evolutionary algorithms to discover control systems that can handle both the complexity of the robot snake itself and also ensure that the artistic properties of the sculpture are fulfilled.

A realization of Trans<->Former \# 13 may include the following milestones:

1: Making a software model of the snake to test different methods The software model may include graphic together with a physical model for the snake. The graphic model can give visual feedback to the artist for evaluation. A physical model can be used in evaluation phase of the evolutionary algorithm. As such, there are at least two goals for the modelling of the snake.

2: Applying evolutionary algorithms to build the control system of the joints. At present time a cellular computation model of the snake is considered. The strongest candidate for a evolutionary algorithm is a cellular programing approach.

3: Evaluation of the control system both for artist purpose and for the purpose of building a scale prototype.

4: Building a scaled prototype.

5: Apply the algorithms and control systems to the prototype. Revise to a real physical model.

6: Building a full scale prototype.

7: making Trans<->Former \# 13 ready for display.

To build the prototype for both the scale and the full scale snake we have planned to use a network solution based on TCP-IP. As such, the control system can be running on a host computer that communicate with each joint sending control signals and receiving status. This approach opens for scalable computation power in each joint as the host computer can be anything from a simple laptop to a supercomputer. Another advantage of the network approach is the possibility to reuse the control system (electrical circuits) in the full scale prototype. This is possible since the actual control system runs on the host computer, the control system in each joint only send and receive signals. The only parts that differ are motors and motor driving circuits.

Point one to three have a a relative known time frame. A pre-project for a master theses is under work autumn 2006. The master theses will be carried out spring 2007. Hopefully a new pre-project autumn 2007 and master theses spring 2008 can be carried out for point four and five. We hope to be able to include students from industrial designer already now to find materials and physical design for the joints.

The project have so far resulted in two master theses at the department of cybernetics.

ASCENDING # 7 
360 x 40 x 40 cm / laquered steel
Trøndelag Senter for Samtidskunst
Trøndelag Centre for Contemporary Arts / 2007

Commissioned by Fokus Bank, Trondheim.

FACE 2 FACE

Stainless steel /
The Petroleum Technical Center, NTNU, Trondheim / 1993