Ambient Field System
Location
Domestic Residency
Year
2024
An exploration of distributed intelligence across domestic space—not a smart home, but a sensitive surface that responds without commands.
The Ambient Field System rejects the notion of the "smart home" entirely. Rather than discrete devices awaiting voice commands, we prototyped a distributed sensing architecture that treats the domestic environment as a single responsive field.
Using pressure-sensitive flooring, thermal gradients, and micro-gesture recognition, the system learns inhabitant patterns without explicit training. There are no screens, no apps, no setup process. The field simply observes, adapts, and responds—adjusting light, temperature, and acoustic properties based on behaviour patterns it detects across weeks of quiet observation.
Through our Domestic Residency program, three households lived with the system for six months. The most consistent feedback: "I forgot it was there, but I noticed when it was gone." This is the benchmark for ambient intelligence—technology you sense only in its absence.
System Architecture
The Field operates on three principles: sense everything, store nothing, respond minimally. Pressure-sensitive flooring tracks movement without recording position. Thermal sensors detect presence without identifying individuals. The system knows someone is in the kitchen making tea, but it doesn't know who, when, or why. It just adjusts the environment accordingly.
This architectural decision—ephemeral sensing without persistent storage—addresses the surveillance anxiety inherent in ambient systems. The Field has no memory beyond immediate context. It learns patterns but forgets specifics. It's attentive without being invasive.
Residency Findings
Across three households (single occupant, couple, family with children), the system achieved 87% environmental optimization within 45 days. But the metric that mattered wasn't optimization—it was invisibility. Exit interviews revealed participants couldn't articulate what the system did, only that spaces "felt better" when it was active.
One participant described it perfectly: "It's like the house learned to breathe with us." This is the design goal for ambient intelligence—not efficiency, but environmental coherence.
Next Project