SARATH THARAYIL
SARATH THARAYILHS.T.WRITEUPSWWRITEUPSCONCEPTSKCONCEPTSPROJECTSPPROJECTSLABLLABABOUTAABOUT
മ
/ SYSTEM

Building thoughtful software, writing notes, and shipping experiments across data, AI, and the web.

No cookies, no tracking. Preferences are stored locally in your browser. Anonymous view counts are kept server-side.

© 2026 Sarath Tharayil/IST --:--:--
← LAB

Reynolds' Boids

May 11, 2026

Craig Reynolds' 1986 flocking model. Each boid follows three rules: stay separated, align with neighbours, move toward the group. Together they produce lifelike swarm behaviour with no central control.

BOIDS120
|
VISION60
|
SPACE28

Click to scatter nearby boids. VISION = perception radius. SPACE = separation distance.

/ NOTES
HOW IT WORKS

Each boid knows only its immediate neighbours. Three steering forces are blended every frame. No leader, no central control — the flock is entirely emergent from local interactions.

THE THREE RULES
SeparationSteer away from neighbours that are too close.
AlignmentMatch the average heading of nearby neighbours.
CohesionMove toward the average position of the group.
WHAT TO TRY

High VISION forms large coordinated flocks. Low SPACE bunches them tight. Click to scatter and watch the flock reform. Low count reveals individual steering in real time.