Unverified50% confidenceBenchmarkExact time

The trained policy can reduce the drone's median lap time to 5.7 seconds, while the time-optimal planner TOGT achieves 4.72 seconds on the same drone, meaning the RL policy is approximately 20% short of the theoretical optimum.

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

10/11/2026

First Seen

Valid until: 1/9/2027

Sources

PPO Reinforcement Learning for Drone Racing: 5.7-Second Lap Times and the Takeoff Problem

Related Entities

TOGTproduct PPOconcept

Related Claims

UnverifiedThe typical speed range of small turbojet-powered drones is 500–700 kilometers per hour; compared to a propeller drone at 185 kilometers per hour, the interception time window over an equivalent distance may be compressed to less than one-third.75% similar UnverifiedCurrent mainstream consumer-grade quadrotor drones (such as the DJI Mini series) generally have a flight endurance of 30–40 minutes, with approximately 60–70% of energy consumption used to counteract gravity and maintain hover altitude.69% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/996323

API

curl https://kongchang.com/api/v1/knowledge/claims/996323

MCP

get_claim(id=996323)