Continuous-time estimation for dynamic obstacle tracking

Ushani, Arash K.; Carlevaris-Bianco, Nicholas; Cunningham, Alexander; Galceran, Enric; Eustice, Ryan M.

doi:10.1109/iros.2015.7353513

Cited by 10 publications

(11 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the SOT algorithms specialize at providing more subtle state information with even centimeter level accuracy. This property is important for many applications in autonomous driving, We further visualize the quality of SOT with the aggregated point clouds, following [15], [31]. As is demonstrated, our algorithm can provide accurate tracking and shape aggregation results.…”

Section: State Estimationmentioning

confidence: 88%

“…[7], [5], [16] jointly use LiDAR and camera for this task. [15], [31] are closest to our objective in using LiDAR only: Held et al [15] aggregates the observed point clouds, while Ushani et al [31] directly optimizes a point cloud model.…”

Section: Point Cloud Completionmentioning

confidence: 94%

“…The only work we found falls into our setting is Ushani et al [31]. Dating back to that period, there is no large-scale dataset fits the evaluation needs, thus [31] only tests on a private dataset, which makes the comparisons with their work infeasible. This is just the reason to propose our new LiDAR-SOT benchmark, which aims to promote the studies in this area.…”

Section: B State Estimation and Tracking In Point Cloudmentioning

confidence: 99%

“…1) Vehicle Shape Visualization: Following the approach in [15], [31], we visualize the aggregated point clouds in Fig. 4 to indicate the quality of tracking.…”

Section: B Analysesmentioning

confidence: 99%

See 3 more Smart Citations

Model-free Vehicle Tracking and State Estimation in Point Cloud Sequences

Pang¹,

Li²,

Wang³

2021

Preprint

View full text Add to dashboard Cite

Estimating the states of surrounding traffic participants stays at the core of autonomous driving. In this paper, we study a novel setting of this problem: model-free singleobject tracking (SOT), which takes the object state in the first frame as input, and jointly solves state estimation and tracking in subsequent frames. The main purpose for this new setting is to break the strong limitation of the popular "detection and tracking" scheme in multi-object tracking. Moreover, we notice that shape completion by overlaying the point clouds, which is a by-product of our proposed task, not only improves the performance of state estimation but also has numerous applications. As no benchmark for this task is available so far, we construct a new dataset LiDAR-SOT and corresponding evaluation protocols based on the Waymo Open dataset [28]. We then propose an optimization-based algorithm called SOTracker based on point cloud registration, vehicle shapes, and motion priors. Our quantitative and qualitative results prove the effectiveness of our SOTracker and reveal the challenging cases for SOT in point clouds, including the sparsity of LiDAR data, abrupt motion variation, etc. Finally, we also explore how the proposed task and algorithm may benefit other autonomous driving applications, including simulating LiDAR scans, generating motion data, and annotating optical flow. The code and protocols for our benchmark and algorithm are available at https://github.com/TuSimple/LiDAR_SOT/. A video demonstration is at https://www.youtube.com/watch? v=BpHixKs91i8.

show abstract

Section: State Estimationmentioning

confidence: 88%

Section: Point Cloud Completionmentioning

confidence: 94%

Section: B State Estimation and Tracking In Point Cloudmentioning

confidence: 99%

“…1) Vehicle Shape Visualization: Following the approach in [15], [31], we visualize the aggregated point clouds in Fig. 4 to indicate the quality of tracking.…”

Section: B Analysesmentioning

confidence: 99%

See 2 more Smart Citations

Model-free Vehicle Tracking and State Estimation in Point Cloud Sequences

Pang¹,

Li²,

Wang³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In this work, we explicitly focus on the state estimation task and assume that association between the point segments (i.e., point segment identity) is given. Instead of performing direct optimization [16,43,53,2], we propose a learning-based approach, that optimizes weights of a neural network, such that, given two LiDAR point clouds, the error in the estimated relative motion is minimized. Such an approach has the advantage that it can improve its performance with an increasing amount of training data.…”

Section: Introductionmentioning

confidence: 99%

AlignNet-3D: Fast Point Cloud Registration of Partially Observed Objects

GroB

Ošep

Leibe

2019

2019 International Conference on 3D Vision (3DV)

View full text Add to dashboard Cite

Methods tackling multi-object tracking need to estimate the number of targets in the sensing area as well as to estimate their continuous state. While the majority of existing methods focus on data association, precise state (3D pose) estimation is often only coarsely estimated by approximating targets with centroids or (3D) bounding boxes. However, in automotive scenarios, motion perception of surrounding agents is critical and inaccuracies in the vehicle close-range can have catastrophic consequences. In this work, we focus on precise 3D track state estimation and propose a learning-based approach for object-centric relative motion estimation of partially observed objects. Instead of approximating targets with their centroids, our approach is capable of utilizing noisy 3D point segments of objects to estimate their motion. To that end, we propose a simple, yet effective and efficient network, AlignNet-3D, that learns to align point clouds. Our evaluation on two different datasets demonstrates that our method outperforms computationally expensive, global 3D registration methods while being significantly more efficient. We make our data, code, and models available at https://www.vision. rwth-aachen.de/page/alignnet.

show abstract

A Probability-Based Artificial Potential Field for Autonomous Vehicles in Avoiding Uncertain Obstacles

Tam

Ánh²

2022

Computational Intelligence Methods for Green Technology and Sustainable Development

View full text Add to dashboard Cite

Continuous-time estimation for dynamic obstacle tracking

Cited by 10 publications

References 19 publications

Model-free Vehicle Tracking and State Estimation in Point Cloud Sequences

Model-free Vehicle Tracking and State Estimation in Point Cloud Sequences

AlignNet-3D: Fast Point Cloud Registration of Partially Observed Objects

A Probability-Based Artificial Potential Field for Autonomous Vehicles in Avoiding Uncertain Obstacles

Contact Info

Product

Resources

About