Joint spatial- and Doppler-based ego-motion estimation for automotive radars

Barjenbruch, Michael; Kellner, Dominik; Klappstein, Jens; Dickmann, Juergen; Dietmayer, Klaus

doi:10.1109/ivs.2015.7225789

Cited by 40 publications

(23 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The sensor velocity, v s = (v x , v y ), mounting position, (l, b), and viewing direction, θ s , are used as dummy variables without ground truths; they are used for regularization to assist training and reduce over-fitting. As previously examined [9]- [11], [13], [40], the observed static target has a velocity that is directly opposite to the sensor velocity, v s . From the perspective of radars measuring the instantaneous radial velocity, v r,i , of targets, the sensor velocity can be estimated using the bearing angle, θ i , and the radial velocity relationships of two or more static targets, as follows:…”

Section: Geometric Constraintsmentioning

confidence: 63%

“…This work, which determined an ego-velocity vector of 2 degrees of freedom (DoF), was extended to the case of multiple distributed radars that deals with the full 2D vehicle motion state, i.e., 3 DoF [10]. A probabilistic approach incorporating spatial registrations of radar scans was also proposed [11]. This joint spatial and Doppler-based estimation functions without lever-arm offsets or motion assumptions but involves significant computational costs.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Complex-Valued Channel Attention and Application in Ego-Velocity Estimation With Automotive Radar

et al. 2021

View full text Add to dashboard Cite

Attention mechanisms have been widely integrated with various neural networks to boost performance. However, when an attention mechanism was applied to a radar ego-velocity estimation network, the importance of carefully handling the amplitude and phase of complex-valued tensor was revealed. Therefore, in this study, we present a self-attention mechanism designed to handle complexvalued tensors in order to capture the rich contextual relationships implied within amplitude and phase. To exploit the advantages of complex-valued attention (CA), we evaluated its impact while performing egovelocity estimation tasks based on radar data, whose amplitude and phase are related to the electromagnetic scattering of the target being observed. Radars are suitable sensors for such tasks as they are capable of long-range detection and instantaneous velocity measurement under variable weather and lighting conditions. In particular, we coupled our CA module with complex-valued neural networks, known to be particularly powerful for handling wave phenomena. The proposed method exhibits robust estimation performance, regardless of whether the Doppler ambiguity problem occurs and eliminates the dependence on preprocessing stages, including target detection and static target indication. Furthermore, it achieves improved stability during training via geometrical constraint regularization, and implicitly allows velocity conversion between the sensor and vehicle frames, even if the mount information of the sensor was not provided. Finally, ablation experiments conducted on extensive real-world datasets show noticeable improvement in estimation performance. INDEX TERMS Automotive radar; complex-valued attention; complex-valued neural network; deep learning; ego-velocity estimation This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.

show abstract

Section: Geometric Constraintsmentioning

confidence: 63%

Section: Related Workmentioning

confidence: 99%

Complex-Valued Channel Attention and Application in Ego-Velocity Estimation With Automotive Radar

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Our motivation behind the second simulated evaluation stems from ego-motion estimation using sensors that measure sets of spatial points, such as radars or lidars. This task can be interpreted as a robust point set registration problem, as shown for example by [17] with radar sensors. We apply the concept from [18] to represent two set of points with Gaussian mixture models and register them in a distributionto-distribution manner.…”

Section: B Point Set Registrationmentioning

confidence: 99%

Advancing Mixture Models for Least Squares Optimization

Pfeifer

Lange

Protzel

2021

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Gaussian mixtures are a powerful and widely used tool to model non-Gaussian estimation problems. They are able to describe measurement errors that follow arbitrary distributions and can represent ambiguity in assignment tasks like point set registration or tracking. However, using them with common least squares solvers is still difficult. Existing approaches are either approximations of the true mixture or prone to convergence issues due to their strong nonlinearity. We propose a novel least squares representation of a Gaussian mixture, which is an exact and almost linear model of the corresponding log-likelihood. Our approach provides an efficient, accurate and flexible model for many probabilistic estimation problems and can be used as cost function for least squares solvers. We demonstrate its superior performance in various Monte Carlo experiments, including different kinds of point set registration. Our implementation is available as open source code for the state-of-the-art solvers Ceres and GTSAM.

show abstract

“…When used between consecutive scans this can provide a very accurate replacement to wheel-based or inertial sensor odometry. In [14] Barjenbruch et. al showed impressive results when estimating velocity and yaw-rate using a similar radar sensor to the ones used in this work.…”

Section: Related Workmentioning

confidence: 99%

“…When localizing we want to remove measurements from moving targets from both X and Y . By thresholding the difference between the measured Doppler velocityṙ i with the expected Doppler velocity for stationary targets, V i , [14] we can remove most moving targets from consideration. V i is the Doppler velocity that would be measured from a stationary target based on the motion of the sensor, where the sensor pose relative to the vehicle pose, s t , is (x s , y s , α s )…”

Section: B Sensor Modelmentioning

confidence: 99%

Vehicle localization with low cost radar sensors

Ward

Folkesson

2016

2016 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

This is the published version of a paper presented at Intelligent Vehicles Symposium (IV), 2016 IEEE. Citation for the original published paper:Ward, E., Folkesson, J. (2016) Vehicle localization with low cost radar sensors. Abstract-Autonomous vehicles rely on GPS aided by motion sensors to localize globally within the road network. However, not all driving surfaces have satellite visibility. Therefore, it is important to augment these systems with localization based on environmental sensing such as cameras, lidar and radar in order to increase reliability and robustness. In this work we look at using radar for localization. Radar sensors are available in compact format devices well suited to automotive applications. Past work on localization using radar in automotive applications has been based on careful sensor modeling and Sequential Monte Carlo, (Particle) filtering. In this work we investigate the use of the Iterative Closest Point, ICP, algorithm together with an Extended Kalman filter, EKF, for localizing a vehicle equipped with automotive grade radars. Experiments using data acquired on public roads shows that this computationally simpler approach yields sufficiently accurate results on par with more complex methods.

show abstract

Joint spatial- and Doppler-based ego-motion estimation for automotive radars

Cited by 40 publications

References 10 publications

Complex-Valued Channel Attention and Application in Ego-Velocity Estimation With Automotive Radar

Complex-Valued Channel Attention and Application in Ego-Velocity Estimation With Automotive Radar

Advancing Mixture Models for Least Squares Optimization

Vehicle localization with low cost radar sensors

Contact Info

Product

Resources

About