| 1. |
- Berlin, Jonas, et al.
(författare)
-
Trajectory Generation for Mobile Robots in a Dynamic Environment using Nonlinear Model Predictive Control
- 2021
-
Ingår i: IEEE International Conference on Automation Science and Engineering. - 2161-8070 .- 2161-8089. ; 2021-August, s. 942-947
-
Konferensbidrag (refereegranskat)abstract
- This paper presents an approach to collision-free, long-range trajectory generation for a mobile robot in an industrial environment with static and dynamic obstacles. For the long-range planning a visibility graph together with A is used to find a collision-free path with respect to the static obstacles. This path is used as a reference path to the trajectory planning algorithm that in addition handles dynamic obstacles while complying with the robot dynamics and constraints. A Nonlinear Model Predictive Control (NMPC) solver generates a collision-free trajectory by staying close to the initial path but at the same time obeying all constraints. The NMPC problem is solved efficiently by leveraging the new numerical optimization method Proximal Averaged Newton for Optimal Control (PANOC). The algorithm was evaluated by simulation in various environments and successfully generated feasible trajectories spanning hundreds of meters in a tractable time frame.
|
|
| 2. |
- de Jong, Roelof S., et al.
(författare)
-
4MOST-4-metre Multi-Object Spectroscopic Telescope
- 2014
-
Ingår i: Ground-based and Airborne Instrumentation for Astronomy V. - : SPIE. - 0277-786X .- 1996-756X. ; 9147
-
Konferensbidrag (refereegranskat)abstract
- 4MOST is a wide-field, high-multiplex spectroscopic survey facility under development for the VISTA telescope of the European Southern Observatory (ESO). Its main science drivers are in the fields of galactic archeology, high-energy physics, galaxy evolution and cosmology. 4MOST will in particular provide the spectroscopic complements to the large area surveys coming from space missions like Gaia, eROSITA, Euclid, and PLATO and from ground-based facilities like VISTA, VST, DES, LSST and SKA. The 4MOST baseline concept features a 2.5 degree diameter field-of-view with similar to 2400 fibres in the focal surface that are configured by a fibre positioner based on the tilting spine principle. The fibres feed two types of spectrographs; similar to 1600 fibres go to two spectrographs with resolution R> 5000 (lambda similar to 390-930 nm) and similar to 800 fibres to a spectrograph with R> 18,000 (lambda similar to 392-437 nm & 515-572 nm & 605-675 nm). Both types of spectrographs are fixed-configuration, three-channel spectrographs. 4MOST will have an unique operations concept in which 5 year public surveys from both the consortium and the ESO community will be combined and observed in parallel during each exposure, resulting in more than 25 million spectra of targets spread over a large fraction of the southern sky. The 4MOST Facility Simulator (4FS) was developed to demonstrate the feasibility of this observing concept. 4MOST has been accepted for implementation by ESO with operations expected to start by the end of 2020. This paper provides a top-level overview of the 4MOST facility, while other papers in these proceedings provide more detailed descriptions of the instrument concept[1], the instrument requirements development[2], the systems engineering implementation[3], the instrument model[4], the fibre positioner concepts[5], the fibre feed[6], and the spectrographs[7].
|
|
| 3. |
- Hess, Georg, 1996, et al.
(författare)
-
LidarCLIP or: How I Learned to Talk to Point Clouds
- 2024
-
Ingår i: Proceedings of the 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). - 2642-9381. - 9798350318920 ; , s. 7423-7432
-
Konferensbidrag (refereegranskat)abstract
- Research connecting text and images has recently seen several breakthroughs, with models like CLIP, DALL•E 2, and Stable Diffusion. However, the connection between text and other visual modalities, such as lidar data, has received less attention, prohibited by the lack of text-lidar datasets. In this work, we propose LidarCLIP, a mapping from automotive point clouds to a pre-existing CLIP embedding space. Using image-lidar pairs, we supervise a point cloud encoder with the image CLIP embeddings, effectively relating text and lidar data with the image domain as an intermediary. We show the effectiveness of Lidar-CLIP by demonstrating that lidar-based retrieval is generally on par with image-based retrieval, but with complementary strengths and weaknesses. By combining image and lidar features, we improve upon both single-modality methods and enable a targeted search for challenging detection scenarios under adverse sensor conditions. We also explore zero-shot classification and show that LidarCLIP outperforms existing attempts to use CLIP for point clouds by a large margin. Finally, we leverage our compatibility with CLIP to explore a range of applications, such as point cloud captioning and lidar-to-image generation, without any additional training. Code and pre-trained models at github.com/atonderski/lidarclip.
|
|
| 4. |
- Hess, Georg, 1996, et al.
(författare)
-
Masked Autoencoder for Self-Supervised Pre-Training on Lidar Point Clouds
- 2023
-
Ingår i: Proceedings - 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2023. ; , s. 350-359
-
Konferensbidrag (refereegranskat)abstract
- Masked autoencoding has become a successful pretraining paradigm for Transformer models for text, images, and, recently, point clouds. Raw automotive datasets are suitable candidates for self-supervised pre-training as they generally are cheap to collect compared to annotations for tasks like 3D object detection (OD). However, the development of masked autoencoders for point clouds has focused solely on synthetic and indoor data. Consequently, existing methods have tailored their representations and models toward small and dense point clouds with homogeneous point densities. In this work, we study masked autoencoding for point clouds in an automotive setting, which are sparse and for which the point density can vary drastically among objects in the same scene. To this end, we propose Voxel-MAE, a simple masked autoencoding pre-training scheme designed for voxel representations. We pre-train the backbone of a Transformer-based 3D object detector to reconstruct masked voxels and to distinguish between empty and non-empty voxels. Our method improves the 3D OD performance by 1.75 mAP points and 1.05 NDS on the challenging nuScenes dataset. Further, we show that by pre-training with Voxel-MAE, we require only 40 of the annotated data to outperform a randomly initialized equivalent. Code is available at https://github.com/georghess/voxel-mae.
|
|
| 5. |
- Hess, Georg, 1996, et al.
(författare)
-
Object Detection as Probabilistic Set Prediction
- 2022
-
Ingår i: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). - Cham : Springer Nature Switzerland. - 1611-3349 .- 0302-9743. - 9783031200793 ; 13670:XVIII, s. 550-566
-
Konferensbidrag (refereegranskat)abstract
- Accurate uncertainty estimates are essential for deploying deep object detectors in safety-critical systems. The development and evaluation of probabilistic object detectors have been hindered by shortcomings in existing performance measures, which tend to involve arbitrary thresholds or limit the detector’s choice of distributions. In this work, we propose to view object detection as a set prediction task where detectors predict the distribution over the set of objects. Using the negative log-likelihood for random finite sets, we present a proper scoring rule for evaluating and training probabilistic object detectors. The proposed method can be applied to existing probabilistic detectors, is free from thresholds, and enables fair comparison between architectures. Three different types of detectors are evaluated on the COCO dataset. Our results indicate that the training of existing detectors is optimized toward non-probabilistic metrics. We hope to encourage the development of new object detectors that can accurately estimate their own uncertainty. Code at https://github.com/georghess/pmb-nll.
|
|
| 6. |
|
|
| 7. |
- Pinto, Juliano, 1990, et al.
(författare)
-
Next Generation Multitarget Trackers: Random Finite Set Methods vs Transformer-based Deep Learning
- 2021
-
Ingår i: Proceedings of 2021 IEEE 24th International Conference on Information Fusion, FUSION 2021. ; , s. 1059-1066
-
Konferensbidrag (refereegranskat)abstract
- Multitarget Tracking (MTT) is the problem of tracking the states of an unknown number of objects using noisy measurements, with important applications to autonomous driving, surveillance, robotics, and others. In the model-based Bayesian setting, there are conjugate priors that enable us to express the multi-object posterior in closed form, which could theoretically provide Bayes-optimal estimates. However, the posterior involves a super-exponential growth of the number of hypotheses over time, forcing state-of-the-art methods to resort to approximations for remaining tractable, which can impact their performance in complex scenarios. Model-free methods based on deep-learning provide an attractive alternative, as they can, in principle, learn the optimal filter from data, but to the best of our knowledge were never compared to current state-of-the-art Bayesian filters, specially not in contexts where accurate models are available. In this paper, we propose a high-performing deeplearning method for MTT based on the Transformer architecture and compare it to two state-of-the-art Bayesian filters, in a setting where we assume the correct model is provided. Although this gives an edge to the model-based filters, it also allows us to generate unlimited training data. We show that the proposed model outperforms state-of-the-art Bayesian filters in complex scenarios, while matching their performance in simpler cases, which validates the applicability of deep-learning also in the model-based regime. The code for all our implementations is made available at https://github.com/JulianoLagana/MT3.
|
|
| 8. |
|
|
