Jinsi Data za Pokémon Go Zilivyofundisha AI Inayowezesha Droni za Kijeshi Sasa

Seti kubwa za data zinazozalishwa na mamilioni ya wachezaji wa Pokémon Go zinapata matumizi mapya yenye hatari kubwa katika teknolojia ya ulinzi. Kupitia mifumo ya AI ya kijiografia ya Niantic, skani za hiari za uhalisia ulioongezwa (AR) zinatumika upya ili kuruhusu droni zinazojiendesha kusafiri bila kuhitaji GPS.

Kutoka Michezo ya Simu hadi Uendeshaji Bila GPS

Mpito kutoka michezo hadi ulinzi ulianza na sasisho la mwaka 2021 la Pokémon Go, ambapo Niantic ilianzisha motisha za ndani ya mchezo kwa wachezaji ili kuskani mazingira ya ulimwengu halisi. Harakati hii iliwageuza mamilioni ya watumiaji kuwa wakusanyaji wa data wa hiari, wakizalisha mabilioni ya pointi za data za ramani za picha zinazojumuisha barabara, majengo, bustani, na miti.

Akiba hii kubwa ya data za picha za usawa wa ardhi ilitumiwa kufundisha mifumo ya msingi ya Niantic. Kufuatia kuanzishwa kwa Niantic Spatial kama kampuni inayojitegemea, teknolojia hii imevuka mipaka ya burudani. Kampuni hiyo imeingia katika ushirikiano wa kimkakati na Vantor, kampuni ya ujasusi wa ulinzi ya Marekani, ili kuunganisha Visual Positioning System (VPS) ya Niantic na programu ya Raptor ya Vantor.

Kutatua Tatizo la Kuzuia na Kuiga Mawimbi ya GPS (Jamming and Spoofing)

Katika maeneo ya migogoro ya kisasa, kama vile yanayoonekana nchini Ukraine na Iran, kuzuia (jamming) na kuiga (spoofing) mawimbi ya GPS hutumiwa mara kwa mara ili kuchanganya droni za kamikaze, UAV za upelelezi, na makombora yanayoongozwa na GPS. Mawimbi ya satelaiti yanapovurugwa, mifumo inayojiendesha kwa kawaida hupoteza uwezo wake wa kusafiri kwa usahihi.

Mchanganyiko wa ramani za 3D za usawa wa ardhi za Niantic Spatial na picha za satelaiti za miaka 20 za Vantor unaunda mfumo wa uratibu wa pamoja unaoruhusu mashine kujitambua kwa kutumia picha pekee. Uendeshaji huu wa "bila GPS" haushindwi na vifaa vya kawaida vya kuzuia mawimbi (signal jammers). Majaribio ya awali ya kiufundi yaliripoti matokeo ya kushangaza, ikiwa ni pamoja na upunguzaji wa makosa kwa hadi 70% na usahihi wa uendeshaji wa takriban mita 1.5.

Mazingira ya Biashara na Ulinzi

Athari za teknolojia hii zinaenea hadi kwenye mikataba mikubwa ya kijeshi. Mnamo Februari 2026, Vantor ilipata mkataba wa Jeshi la Marekani wenye thamani ya hadi dola milioni 217 kwa ajili ya programu ya One World Terrain. Programu hii inalenga kujenga data za mazingira ya 3D zenye usahihi wa juu kwa ajili ya Synthetic Training Environment ya Jeshi, ambayo hutumiwa kwa mazoezi na simulizi za misheni halisi.

While Niantic has clarified that raw player scans were not handed directly to defense contractors—noting that the data was used to train general foundation models via opt-in user permissions—the lineage of the technology is clear. The corporate structure has also shifted to reflect this specialization: Niantic’s gaming division was acquired by Scopely for $3.5 billion in March 2025, while Niantic Spatial operates as a standalone entity focused entirely on geospatial AI.

Why This Matters for the AI Industry

This development highlights a growing trend in AI: the "dual-use" nature of massive consumer datasets. Data collected for low-stakes consumer engagement (like catching digital monsters) is becoming the bedrock for high-stakes industrial and military autonomy. As spatial AI matures, the ability to turn visual pixels into precise, navigable 3D coordinates will be the defining factor in the next generation of autonomous robotics.

Key Takeaways

  • Data Repurposing: Billions of ground-level visual scans from Pokémon Go players were used to train Niantic’s spatial AI foundation models.
  • GPS-Free Autonomy: The technology allows drones and vehicles to navigate using visual positioning, making them immune to GPS jamming and spoofing.
  • Proven Accuracy: Integration with Vantor’s Raptor software has demonstrated a 70% reduction in error, achieving a navigation accuracy of 1.5 meters.