Motion blur at closing speed
A frame sensor integrating over milliseconds smears a crossing target into a streak precisely when the guidance loop needs a centroid. Event pixels report change in microseconds and never integrate.
Event-Based Vision · Edge Autonomy
The engagement is decided in the terminal second.
Balor Defense builds event-based seeker payloads and edge-AI guidance for the last few hundred meters of a counter-UAS engagement — where frame cameras blur, saturate, and run out of compute budget.
TRL 5/6Validated at sea, 2026
Navy · Army · USSOCOMSBIR programs led
7,000 ft²In-house design & fab
13 acresPrivate test range
The Problem
Why the last 500 meters breaks conventional imaging
Every failure mode below compounds at exactly the moment the engagement is resolved.
Sun, sky and glare
Conventional imagers give up the target against a bright sky or a low sun. Per-pixel logarithmic response holds contrast across a dynamic range no global-shutter frame camera matches.
Compute you cannot carry
Dense frames force a GPU-class processor into a package that has neither the mass budget nor the power for one. A sparse event stream puts useful tracking inside an embedded envelope.
Latency you cannot spend
Detect-to-command budgets in the terminal phase are measured in single-digit milliseconds. Pipelines built around frame cadence spend that budget before inference begins.
Proof
Not a rendering
In May 2026 we ran Balor from a U.S. Navy vessel underway at ONR's Wave Breaker exercise in Key West — detection and on-the-move tracking of small UAS from a moving deck. TRL 5/6, in a real environment, against real targets.
The clip is an unedited field recording from our own range. The overlay is the module's live on-board output.
What We Build
Three layers, delivered as one integrated chain
Sensor, estimator and effector interface are designed together. A seeker that is excellent in isolation is worth nothing if the handoff to the weapon is where the latency hides.
Seeker payloads
Event-based and dual-modality sensor heads sized for interceptor and effector integration — optics, sensor, processing and interface as one qualified assembly.
Guidance & fire control
Track formation, state estimation and command generation running on the airframe or the mount. EKF and visual-inertial estimation tuned for small, agile, low-signature targets.
Edge autonomy stack
The software that makes the sensor useful: event clustering, track association, classification and handoff, running deterministically on FPGA and embedded SoC targets.
Application Domains
Built for the close fight
C-UAS interceptors
Terminal seeking for kinetic interceptors engaging Group 1–3 UAS, where the target is small, cold, maneuvering, and often silhouetted against sky.
Small arms fire control
Aided-aim and lead-angle computation for dismounted and crew-served weapons, with sensing that survives muzzle disturbance and rapid slew.
Point defense
Fixed and mobile site protection — cueing, track custody and effector handoff for close-in layers where reaction time is the binding constraint.
Lineage
A Paladin Robotics line
Balor is the productization of an event-based vision and edge-autonomy capability developed inside Paladin Robotics across multiple SBIR programs for the Navy, the Army and USSOCOM. The underlying work is not a technology demonstration looking for a mission — it was built against the mission first.
Paladin is a veteran-owned defense robotics R&D firm operating its own manufacturing space and private test grounds. Development, integration and flight test happen under one roof.
- Principal investigator on Navy, Army and USSOCOM sensing SBIR programs
- Event-camera and GigE Vision sensor engineering in-house
- FPGA, embedded firmware and mechanical design under one team
- Balor validated at sea at ONR Wave Breaker, May 2026 — TRL 5/6
- 13-acre private range for captive-carry and flight test iteration
- SDVOSB certified and SAM.gov registered
Latest
News & program updates
Paladin Robotics launches Balor Defense
Paladin's event-based vision and edge-autonomy work is now a dedicated product line focused on seeker and guidance solutions for counter-UAS interceptors, small arms and point defense.
Field note: what breaks in the last 500 meters
Four failure modes that compound at precisely the moment a counter-UAS engagement is resolved — and why they are sensor architecture problems, not algorithm problems.
Bring us a hard engagement problem
Program offices, primes and integrators: if you have a terminal-phase sensing problem that conventional EO has not solved, we would like to see it.