Navigating in environments where GPS signals are unavailable, such as tunnels and underground parking, has been a persistent challenge. A team from Wuhan University and Chongqing University has developed a solution, the DMDVDR framework, which uses a deep neural network, AVNet, to estimate a vehicle's position with smartphone inertial sensors. This system merges artificial intelligence with control theory, processing data from a smartphone's IMU to estimate vehicle orientation and velocity, then using an Invariant Extended Kalman Filter to reduce sensor noise and drift, achieving a horizontal translation error of just 0.4% in tests.
This innovation not only improves personal navigation but also has potential applications in autonomous parking assistance and fleet management in GPS-denied areas. Offering a scalable and cost-effective alternative to traditional navigation systems, the DMDVDR framework marks a significant advancement in smart mobility. For more information, visit https://doi.org/10.1186/s43020-025-00168-7.
