Aeva exits stealth, is developing a sensor for self-driving cars that is a cross between lidar, good for measuring distance, and radar, good for measuring speed
PALO ALTO, Calif. — Soroush Salehian raised both arms and spun in circles as if celebrating a touchdown.
Context & Ripple Effects
Aeva's 2017 stealth exit is the opening move of an arc the related coverage traces end to end: founded by engineers out of Apple, the Palo Alto startup bets that fusing lidar's distance measurement with radar's speed measurement beats either alone. Within a year it had productized the idea as a lightweight, low-power box combining LIDAR, motion sensors and cameras for autonomous cars, backed by a $45M Series A.
The bet paid off on paper: by early 2021 Aeva was raising $200M at a ~$3B valuation ahead of a reverse merger, part of a wave that also took rival AEye public via SPAC at $2B. But the coverage also shows the counter-argument gaining ground — camera-first Wayve raised on the thesis that sophisticated sensors like lidar are unnecessary, and years later Teradar claimed a solid-state sensor that could outperform lidar and radar outright.
First-order effects
- Self-driving car developers evaluating sensor suites gain a new option that measures both distance and speed in one unit, directly challenging pure-lidar vendors on capability per box.
- Aeva converts its founders' Apple pedigree into a funded development program, moving from stealth R&D to a competitive position against established lidar makers.
Second-order effects
- Rival lidar startups are pushed toward the same exit path Aeva eventually took — AEye's $2B SPAC shows public markets becoming the funding mechanism for sensor-hardware scale-up.
- The camera-only camp gets a fundraising foil: Wayve's $20M Series A was pitched explicitly against sensor-heavy stacks like Aeva's, making modality choice a differentiator for capital.
Third-order effects
- If hybrid and solid-state challengers keep attracting nine-figure rounds while camera-only approaches win their own backers, autonomous-vehicle perception fragments into competing hardware philosophies rather than converging on one sensor standard.
- Sensor fusion becomes a durable battleground of the autonomy stack: Teradar's 2028-targeted claim to beat lidar and radar suggests each generation of entrants must argue superiority over the previous modality, not just parity.
The trend: Autonomous-vehicle sensing is splintering into rival modalities — fused lidar-radar hybrids, pure lidar, and camera-only stacks — with venture and public-market capital arbitraging between them.