The sunlight problem: why cheap laser sensors fail outdoors
The obvious sensor for this project is a small infrared time-of-flight (ToF) chip — the VL53 family that powers robot vacuums and phone autofocus. Cheap, tiny, accurate. There’s just one problem, and it’s a big one: the sun.
These sensors work by timing reflected pulses of 940 nm infrared light. Sunlight contains enormous amounts of 940 nm light, and a cap-brim sensor aims slightly upward at branches silhouetted against bright sky — geometrically the worst possible case. ST’s own engineers say it plainly on their forums: the VL53 series are effectively indoor sensors; in direct sun the receiver saturates and range can collapse to under a metre. Independent testing (PiWars) confirms the marketing mitigation modes underdeliver in real sun.
The industry already voted
Once you know to look, the pattern is everywhere. Every shipped, affordable assistive wearable — BuzzClip, Sunu Band, iGlasses, WeWALK — chose ultrasound, which doesn’t care about light. Car parking sensors: ultrasonic for decades. Drone altimeter docs (ArduPilot) explicitly mark VL53 sensors “not recommended for outdoor use.” Overheight-vehicle detection systems — this exact problem at road scale — use break-beams and ultrasound, never reflective IR. Even robot lawnmower complaints are informative: ultrasonic sensors false-triggering on grass is their failure mode and this project’s success mode, because foliage reflecting ultrasound is exactly what a branch detector wants.
What this project does about it
Prototype 1 keeps a €9 VL53L0X anyway — for desk and indoor testing, where the haptic-interaction questions (comfort, legibility, thresholds) don’t depend on sunlight at all. Its outdoor degradation gets measured and published, not hidden.
For outdoors, the plan is waterproof ultrasonic probes (the same sealed type used in car bumpers), which are immune to sunlight by physics and shrug off rain. Two are on order for A/B testing.
And for the future, a ranked upgrade ladder, each rung only climbed if test data demands it: the Benewake TF-NOVA (~€55), whose deliberate 14°×1° line-shaped beam is the best thin-branch geometry found at any price under €60 and is rated to full-sun brightness; the tiny Benewake TFS20-L if size becomes critical; and 24/60 GHz radar as the experimental endgame — radar passes through plastic, so the sensor can live fully sealed inside the hat with no aperture at all.
Waterproofness, the axis nobody markets
Hiking means rain and sweat. Ranked by wet-weather robustness: radar first (enclosed by design, best physics in falling rain), sealed ultrasonic probes second (rain on the face is a non-event), and optical sensors last — they need a window, droplets blind them, and airborne rain causes false returns. Another quiet argument for the ultrasonic-first outdoor plan.