Multi-sensor system improves motion tracking in real time
RESPLE2: Inertial-Informed Recursive Non-Uniform Spline Estimation for Multi-Sensor Odometry with Global Refinement
Robotics
Summary
Tracking exact movement is tough when things move quickly or in complicated ways. The authors developed RESPLE2, a system that uses a special math tool called non-uniform B-splines to represent motion precisely with data from multiple sensors like cameras, lasers, and motion detectors. This system predicts and refines the movement path efficiently, even during fast or tricky motions, and can update itself continuously in real time. The authors tested it with various robots and devices, showing it works better or as well as other advanced tracking systems.
What this means in practice
- •For robotics engineers: Track and refine robot positions accurately during complex and rapid movements using fused data from various onboard sensors.
- •For autonomous vehicle developers: Enhance real-time pose estimation for ground and aerial vehicles by combining inertial, visual, and LiDAR inputs through a unified spline-based framework.$Commercial implications: Enables more reliable navigation and control systems that can be marketed for advanced autonomous vehicle platforms.
Authors
Ziyu Cao, William Talbot, Song Li, Pakpong Chirarattananon, Fredrik Gustafsson, Marco Hutter, Kailai Li
Abstract
We present RESPLE2, a recursive multi-sensor state estimation framework using non-uniform B-splines for 6-DoF motion representation. By exploiting the knot-clamping boundary property, we propose a theoretically grounded mechanism for exact control-point prediction directly informed by inertial kinematics along the trajectory, with knot placement adapted to motion and measurement availability. Within an iterated extended Kalman filter, we perform this prediction using a subset of inertial measurements. A subsequent reset step preserves trajectory consistency as the knots extend. The update step further incorporates heterogeneous sensor measurements at their native timestamps through continuous-time observation models, supporting flexible combinations of LiDARs, cameras, inertial measurement units, and leg kinematics. We further develop a lightweight global optimization module that directly refines non-uniform B-spline control points by incorporating loop-closure constraints derived from LiDAR-visual cues. We conduct extensive evaluations on public and experimental datasets across handheld, tracked, legged, and hybrid aerial-legged platforms, covering highly dynamic maneuvers with extreme accelerations. Compared with state-of-the-art multi-sensor odometry systems, RESPLE2 achieves competitive or superior estimation accuracy and robustness in the corresponding sensor configurations, particularly under aggressive motion, while maintaining real-time operation. We publicly release our code and experimental datasets at https://github.com/ASIG-X/RESPLE2.