B-Pro Show 2020
Explore
Autumn Show 2020
Explore



Close
Project sections

Autonomous Transhumance

Project details

Programme
Research Cluster RC1
Group Name Team Earth

A third of all land on earth is composed of grasslands, supporting pastures for nomadic herders and providing crucial ecosystem services. The Nenets of Yamal Peninsula are one such nomadic community that have their own indigenous knowledge system revolving around reindeer herding. Transhumance is their way of life and a key ecological function. Due to a rise in temperature anomalies there has been an emergence of methane mounds known as pingos across the peninsula, and an increase in the extraction of natural gas by corporate giants. These changes have resulted in transmigration routes becoming threatened. The Russian Arctic, once a wasteland, has now become a zone of confluent resources disrupting the once stable ecology and transhumant lifestyles.


Autonomous Transhumance proposes a new way of herding while responding to a warming Arctic by non-invasively territorialising and regenerating the landscape with a family of two autonomous robots. AuRoRaH is a robotic shepherd that guides reindeer to seasonal pastures by triggering their sense of smell. While AuRega is a robotic landscape acupuncturist that regenerates the landscape by relieving methane build up. The project performs through a strategy of local interventions, networking the living and non-living agents in a tight rope act that balances the ecology. The regenerated landforms are visual markers of a changing Arctic due to global climate change.

01

The Arctic Urban

Autonomous Transhumance

The project proposes a family of two autonomous robots, programmed to noninvasively re-tame the land, taking over responsibilities of herding and ecological preservation.

AuRoRaH and AuRega

AuRoRaH and AuRega

AuRoRa is a robotic shepherd that herds reindeer by triggering their sense of smell. AuRega is a robotic landscape acupuncturist that regenerates landscape by relieving methane build up.

Project timeline

Methane Era

Methane Era

In the year 2520, the Arctic urban is dependent on a new invisible currency, altering the primary sense of perception. Methane is the emergent source of energy powering the city.

Episodic Memories

Episodic Memories

Smells are capable of triggering emotions and can store memories. Memories of the Arctic lost due to climate change are documented.

02

Decoding Transhumance

Decoding Arctic

Machine learning algorithms decode the Arctic environment. A new set of herding rules are identified where the system makes decisions based on the changing actor dynamics.

Smell-Based Herding

Smell-Based Herding

Herding simulations are used to create a catalogue of smells and related mixing and herding rules to be offloaded to AuRoRaH.

Decoding Pingo

Decoding Pingo

A convolutional neural network (CNN) model is created for pingo identification and a linear regression model is used for its qualification. Analysis criteria used in the model include: stage, ice, methane, methane amount, reindeer capacity. A new pingo vocabulary is recoded.

Documented Reindeer at Svalbard, Norway

Reindeer are a keystone species of Yamal. They have an extremely keen olfactory sense which assists them in finding food and in navigation.

Herding Rules Simulation

Herding Rules Simulation

The herding model is trained using Unity machine learning agents to attract reindeer using smell attractor logic. The reindeer herd then follows the agents until they reach maximum pingo capacity.

03

Smell Library

Recoding Smells

AuRoRaH mixes smells within its chamber from the scents extracted from materials in the collection tubes. From these mixed smells, AuRoRaH curates a smell catalogue for various situations.

Recoding Landscapes

Recoding Landscapes

AuRega prevents pingo explosions by puncturing and creating earthen mounds on methane leaks. The robotic landscape acupunturist uses adaptive formwork to create different patterns for optimal vegetation growth.

Pattern Tests

AuRega generates different patterns on the pingo to stimulate vegetation growth.

Pingo Monitoring

AuRega uses machine learning to monitor pingos for methane emissions and vegetation growth in real time, relaying information back to AuRoRaH.

Autonomous Transhumance

Autonomous Transhumance

A glance at a speculative future resulting from project Autonomous Transhumance.

Share on , LinkedIn or

Close

Index of Works

The Bartlett
B-Pro & Autumn Shows 2020
27 November – 11 December 2020
Explore