Continuous-time Radar-inertial Odometry for Automotive Radars (2201.02437v1)

Abstract: We present an approach for radar-inertial odometry which uses a continuous-time framework to fuse measurements from multiple automotive radars and an inertial measurement unit (IMU). Adverse weather conditions do not have a significant impact on the operating performance of radar sensors unlike that of camera and LiDAR sensors. Radar's robustness in such conditions and the increasing prevalence of radars on passenger vehicles motivate us to look at the use of radar for ego-motion estimation. A continuous-time trajectory representation is applied not only as a framework to enable heterogeneous and asynchronous multi-sensor fusion, but also, to facilitate efficient optimization by being able to compute poses and their derivatives in closed-form and at any given time along the trajectory. We compare our continuous-time estimates to those from a discrete-time radar-inertial odometry approach and show that our continuous-time method outperforms the discrete-time method. To the best of our knowledge, this is the first time a continuous-time framework has been applied to radar-inertial odometry.

• The paper presents a continuous-time trajectory representation that enables closed-form pose estimation from asynchronous radar and inertial data.
• The paper integrates radar and IMU measurements to enhance ego-motion accuracy, outperforming traditional discrete-time methods.
• The paper demonstrates improved performance in adverse weather, showcasing its potential for advancing reliable autonomous driving systems.

The paper "Continuous-time Radar-inertial Odometry for Automotive Radars" introduces an innovative approach to enhancing ego-motion estimation for automotive vehicles by leveraging a continuous-time framework to fuse data from multiple automotive radars and an inertial measurement unit (IMU). This methodology was motivated by the need for robust sensing solutions under adverse weather conditions, which often impair the performance of camera and LiDAR sensors. Radars, known for their resilience in such conditions and their growing adoption in passenger vehicles, present a promising alternative.

The key contributions of this paper include:

1. Continuous-time Trajectory Representation: Unlike traditional discrete-time approaches which sample sensor data at fixed intervals, the authors utilize a continuous-time representation. This framework ensures more flexible and efficient fusion of asynchronous and heterogeneous sensor data, optimizing pose estimation and allowing for the calculation of poses and their derivatives at any point along the trajectory in closed-form.
2. Enhanced Radar-inertial Odometry: By applying the continuous-time approach, the paper addresses the limitations of discrete-time methods, particularly in handling the asynchronous nature of radar and IMU measurements. The proposed system integrates these measurements seamlessly, leading to more accurate and reliable ego-motion estimates.
3. Comparison and Performance Evaluation: The authors conducted a comparative analysis between their continuous-time radar-inertial odometry approach and a traditional discrete-time method. The results demonstrated that the continuous-time method significantly outperformed the discrete-time counterpart, showcasing better trajectory estimation accuracy.

In essence, this paper not only introduces the continuous-time framework to radar-inertial odometry for the first time but also validates its superiority over discrete-time approaches in terms of accuracy and robustness in various operating conditions. This advancement has potential implications for the development of more reliable autonomous driving systems that can function effectively regardless of weather challenges.