Deep Inference for Covariance Estimation: Learning Gaussian Noise Models for State Estimation

Liu, Katherine; Ok, Kyel; Vega-Brown, William; Roy, Nicholas

doi:10.1109/icra.2018.8461047

Cited by 54 publications

(52 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The effect of unpredictable nonuniform noise as well as external environmental conditions is also inevitable [35]. To enhance the accuracy of localization, the solutions found in the literature can be classified into: (1) controlling the environment under investigation [36], (2) sensor data fusion [37], [38], (3) improving measurement covariane estimation [39], [30], [35], [40], or (4) correcting measurement errors, which can be further classified into classical [41], [42], [43], [44] and learning approaches [16], [45], [34], [46], [17], [47].…”

Section: Enhancing Slam Estimation Accuracymentioning

confidence: 99%

“…Another example of measurement fusion can be found in [38] where measurements recorded by multiple IMUs along with other extroceptive sensors were integrated to improve localization accuracy. Instead of assuming a fixed measurement noise model, the work proposed in [39] predicts the noise model based on raw measurements by means of a DNN. The DNN was able to accurately predict the covariance of measurements obtained by light and vision sensors.…”

Section: Enhancing Slam Estimation Accuracymentioning

confidence: 99%

See 1 more Smart Citation

A Stacked LSTM-Based Approach for Reducing Semantic Pose Estimation Error

Azzam

Alkendi

Taha³

et al. 2021

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

Achieving high estimation accuracy is significant for semantic simultaneous localization and mapping (SLAM) tasks. Yet, the estimation process is vulnerable to several sources of error including limitations of the instruments used to perceive the environment, shortcomings of the employed algorithm, environmental conditions, or other unpredictable noise. In this paper, a novel stacked long short term memory (LSTM) based error reduction approach is developed to enhance the accuracy of semantic SLAM in presence of such error sources. Training and testing datasets were constructed through simulated and realtime experiments. The effectiveness of the proposed approach was demonstrated by its ability to capture and reduce semantic SLAM estimation errors in training and testing datasets. Quantitative performance measurement was carried out using the absolute trajectory error (ATE) metric. The proposed approach was compared to vanilla and bidirectional LSTM networks, shallow and deep neural networks, and to support vector machines. The proposed approach outperforms all other structures and was able to significantly improve the accuracy of semantic SLAM. To further verify the applicability of the proposed approach, it was tested on real-time sequences from the TUM RGB-D dataset, where it was able to improve the estimated trajectories.

show abstract

Section: Enhancing Slam Estimation Accuracymentioning

confidence: 99%

Section: Enhancing Slam Estimation Accuracymentioning

confidence: 99%

A Stacked LSTM-Based Approach for Reducing Semantic Pose Estimation Error

Azzam

Alkendi

Taha³

et al. 2021

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

show abstract

“…The work in [16] presents an inference framework based on a deep ANN, DICE, that can be trained on raw images. This work, to the best of our knowledge, represents the first instance of the use of a ANN to infer the uncertainty of a measurement model in a similar approach to ours.…”

Section: Related Workmentioning

confidence: 99%

PointNetKL: Deep Inference for GICP Covariance Estimation in Bathymetric SLAM

Torroba

Sprague

Bore

et al. 2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Registration methods for point clouds have become a key component of many SLAM systems on autonomous vehicles. However, an accurate estimate of the uncertainty of such registration is a key requirement to a consistent fusion of this kind of measurements in a SLAM filter. This estimate, which is normally given as a covariance in the transformation computed between point cloud reference frames, has been modelled following different approaches, among which the most accurate is considered to be the Monte Carlo method. However, a Monte Carlo approximation is cumbersome to use inside a time-critical application such as online SLAM. Efforts have been made to estimate this covariance via machine learning using carefully designed features to abstract the raw point clouds [1]. However, the performance of this approach is sensitive to the features chosen. We argue that it is possible to learn the features along with the covariance by working with the raw data and thus we propose a new approach based on PointNet [2]. In this work, we train this network using the KL divergence between the learned uncertainty distribution and one computed by the Monte Carlo method as the loss. We test the performance of the general model presented applying it to our target use-case of SLAM with an autonomous underwater vehicle (AUV) restricted to the 2-dimensional registration of 3D bathymetric point clouds.

show abstract

“…Albeit promising, the method obtains large translational error > 30% in their stereo odometry experiment. Finally, [25] uses deep learning for estimating covariance of a local odometry algorithm that is fed into a global optimization procedure, and in [26] we used Gaussian processes to learn a wheel encoders error. Our conference paper [20] contains preliminary ideas, albeit not concerned at all with covariance adaptation: a neural network essentially tries to detect when to perform ZUPT.…”

Section: Introductionmentioning

confidence: 99%

AI-IMU Dead-Reckoning

Brossard

Barrau

Bonnabel

2020

IEEE Trans. Intell. Veh.

179

View full text Add to dashboard Cite

In this paper we propose a novel accurate method for dead-reckoning of wheeled vehicles based only on an Inertial Measurement Unit (IMU). In the context of intelligent vehicles, robust and accurate dead-reckoning based on the IMU may prove useful to correlate feeds from imaging sensors, to safely navigate through obstructions, or for safe emergency stops in the extreme case of exteroceptive sensors failure. The key components of the method are the Kalman filter and the use of deep neural networks to dynamically adapt the noise parameters of the filter. The method is tested on the KITTI odometry dataset, and our dead-reckoning inertial method based only on the IMU accurately estimates 3D position, velocity, orientation of the vehicle and self-calibrates the IMU biases. We achieve on average a 1.10% translational error and the algorithm competes with top-ranked methods which, by contrast, use LiDAR or stereo vision. We make our implementation open-source at: https://github.com/mbrossar/ai-imu-drIndex Terms-localization, deep learning, invariant extended Kalman filter, KITTI dataset, inertial navigation, inertial measurement unit

show abstract

Deep Inference for Covariance Estimation: Learning Gaussian Noise Models for State Estimation

Cited by 54 publications

References 20 publications

A Stacked LSTM-Based Approach for Reducing Semantic Pose Estimation Error

A Stacked LSTM-Based Approach for Reducing Semantic Pose Estimation Error

PointNetKL: Deep Inference for GICP Covariance Estimation in Bathymetric SLAM

AI-IMU Dead-Reckoning

Contact Info

Product

Resources

About