Artificial intelligence news.
Straight from the original announcements.

No team wins DARPA's CIDAR passive ranging challenge, but Zetascope comes closest

September 25, 2026, 14:08Source: DARPA

DARPA's Computational Imaging Detection and Ranging (CIDAR) challenge has ended without a prize winner in the final round, as no team met the performance threshold for passive ranging out to 10 kilometers. The agency says the Zetascope team came closest and describes the outcome as valuable progress.

Key points
  • No team met the performance threshold for a prize in the final round of DARPA's CIDAR challenge.
  • The goal was passive imaging algorithms measuring distances out to 10 kilometers with the accuracy and low latency of active ranging systems.
  • Zetascope came closest to meeting the required metrics, according to DARPA.
  • Final-round prizes of $1 million, $600,000 and $400,000 were on offer, with targets at ranges from 2.5 to 10 kilometers.
  • DARPA says the results clarify what technical advances are still needed.

The goal: active-level accuracy without emitting a signal

The final round asked competitors to develop passive imaging algorithms that can measure distances out to 10 kilometers with the accuracy and low latency of today's active ranging systems. According to DARPA, no team cleared that bar, so none earned a prize in the final round.

The agency set the objective to be deliberately difficult. Jonathan Hoffman, Ph.D, deputy director of DARPA's Multi X Office, said: "CIDAR captures the spirit of DARPA. We take big swings at problems where success is far from guaranteed because that is how we learn what technology can achieve today, and where more work needs to be done."

Why passive ranging matters

DARPA designed CIDAR around a basic limitation in optical ranging. Active systems such as laser detection and ranging and laser range finders can deliver highly accurate distance measurements. They depend on emitted signals, however, and those emissions can reveal the observer's presence.

Passive systems avoid emitting anything, but according to DARPA, current approaches capture only a fraction of the distance information potentially available in images. The challenge asked participants to close that gap using computational imaging, machine learning and artificial intelligence.

How the challenge unfolded

CIDAR was announced in December 2024 and ran through multiple stages. Five small businesses and universities received initial $200,000 prizes. Five teams later earned $50,000 semifinal prizes in January 2026.

In the final round, competitors analyzed a new set of imagery for a chance at prizes of $1 million, $600,000 and $400,000. To qualify, teams had to meet a defined minimum performance threshold across targets at ranges from 2.5 to 10 kilometers.

DARPA said the Zetascope team's performance came closest to meeting the metrics required to win. Its submission did not cross the required threshold, but the agency said the team showed particularly promising progress toward CIDAR's goals.

What the result means

Hoffman said no team cleared the bar, but the competitors "pushed the boundaries of passive ranging and gave us a much clearer understanding of what works, what doesn't, and what still needs to be solved."

DARPA says the challenge helped define the technical advances still needed for accurate, low-latency passive ranging at long distances. It adds that the effort provides insights and a stronger foundation for future work in the field. The source does not announce a follow-on program or timeline.

Related news

Boston Dynamics puts wheels on Atlas to test its limits
ResearchNovember 9, 2026, 15:50

Boston Dynamics puts wheels on Atlas to test its limits

Boston Dynamics has fitted its Atlas humanoid robot with wheels to test how classical motor controls and whole-body reinforcement learning work together under demanding…

Berkeley researchers outline an agentic approach to robot programming
ResearchNovember 9, 2026, 15:00

Berkeley researchers outline an agentic approach to robot programming

UC Berkeley researchers have outlined a robotics approach in which teams of AI agents generate, test and refine robot-control programs in simulation. They call the…

FFR Says OmniMind Tops RoboColiseum Embodied AI Ranking
ResearchNovember 9, 2026, 02:42

FFR Says OmniMind Tops RoboColiseum Embodied AI Ranking

FF EAI Robotics Ecosystem says its OmniMind embodied AI model ranked No. 1 on RoboColiseum's global leaderboard as of November 6. The company also reported sales…