Track Reconstruction Research Articles

Accelerating graph-based tracking tasks with symbolic regression

Abstract The reconstruction of particle tracks from hits in tracking detectors is a computationally intensive task due to the large combinatorics of detector signals. Recent efforts have proven that ML techniques can be successfully applied to the tracking problem, extending and improving the conventional methods based on feature engineering. 
However, complex models can be challenging to implement on heterogeneous trigger systems, integrating architectures such as FPGAs.
Deploying the network on an FPGA is feasible but challenging and limited by its resources. An efficient alternative can employ symbolic regression (SR). We propose a novel approach that uses SR to replace a graph-based neural network. Substituting each network block with a symbolic function preserves the graph structure of the data and enables message passing. 
The technique is perfectly suitable for heterogeneous hardware, as it can be implemented more easily on FPGAs and grants faster execution times on CPU with respect to conventional methods.
While the tracking problem is the target for this work, it also provides a proof-of-principle for the method that can be applied to many use cases.

Real-Time LiDAR–Inertial Simultaneous Localization and Mesh Reconstruction

In this paper, a novel LiDAR–inertial-based Simultaneous Localization and Mesh Reconstruction (LI-SLAMesh) system is proposed, which can achieve fast and robust pose tracking and online mesh reconstruction in an outdoor environment. The LI-SLAMesh system consists of two components, including LiDAR–inertial odometry and a Truncated Signed Distance Field (TSDF) free online reconstruction module. Firstly, to reduce the odometry drift errors we use scan-to-map matching, and inter-frame inertial information is used to generate prior relative pose estimation for later LiDAR-dominated optimization. Then, based on the motivation that the unevenly distributed residual terms tend to degrade the nonlinear optimizer, a novel residual density-driven Gauss–Newton method is proposed to obtain the optimal pose estimation. Secondly, to achieve fast and accurate 3D reconstruction, compared with TSDF-based mapping mechanism, a more compact map representation is proposed, which only maintains the occupied voxels and computes the vertices’ SDF values of each occupied voxels using an iterative Implicit Moving Least Squares (IMLS) algorithm. Then, marching cube is performed on the voxels and a dense mesh map is generated online. Extensive experiments are conducted on public datasets. The experimental results demonstrate that our method can achieve significant localization and online reconstruction performance improvements. The source code will be made public for the benefit of the robotic community.

Spaceborne COTS-Capsule hodoscope: Detecting and characterizing particle radiation

The COTS-Capsule Spaceborne hodoscope was launched into low Earth orbit aboard the International Space Station and operated from 2021 to 2022. The primary objectives of the payload are measuring and characterizing the radiation environment within the space station, serving as a technology demonstrator for the COTS-Capsule radiation mitigation apparatus, and testing the interaction of high-energy cosmic particles on-orbit with our detectors. The payload features a particle hodoscope equipped with an array of novel detectors based on polyvinyl toluene scintillators and silicon photomultiplier sensors that are used for radiation detection and characterization employing 2D intensity-based triangulation. This paper provides a comprehensive account of the COTS-Capsule payload’s construction, preflight performance, testing, qualification, and on-orbit calibration. The hodoscope, sensitive to ionizing cosmic particles, facilitates impinging particle track reconstruction by multi-detector 2D position estimation, energy deposition estimation, and linear energy transfer estimation.

The FLUKA Monte Carlo simulation of the magnetic spectrometer of the FOOT experiment

The FOOT experiment of INFN is devoted to the measurement of the nuclear fragmentation double differential cross sections useful for the improvement of calculation models adopted in hadrontherapy and radioprotection. A detailed Monte Carlo simulation of the FOOT magnetic spectrometer has been implemented in order to optimize the design and to guide data analysis. This task has been accomplished by means of the FLUKA Monte Carlo code. The input files of the FLUKA simulations are created from the software framework of the experiment, in order to have a consistent generation and description of geometry and materials in both simulation and data analysis. In addition, this ensures the possibility of processing both simulated and real data with the same data analysis procedures. Databases containing specific parameters describing the setup employed in each different data taking campaign are used. A customized event-by-event output of the Monte Carlo code has been developed. It can be read out by the general software framework of FOOT, enabling access to the generation history of all particles in the same event. This output structure therefore gives the possibility to perform a detailed analysis and study of all relevant processes, allowing the detailed tracking reconstruction of all individual particles. Examples of results are presented.

Pose Tracking and Object Reconstruction Based on Occlusion Relationships in Complex Environments

Pose Tracking and Object Reconstruction Based on Occlusion Relationships in Complex Environments

Commissioning of the MIGDAL detector with fast neutrons at NILE/ISIS

Many dark matter experiments are exploiting the Migdal effect, a rare atomic process, to improve sensitivity to low-mass (sub-GeV) WIMP-like dark matter candidates. However, this process is yet to be directly observed in nuclear scattering. The MIGDAL experiment aims to make the first unambiguous measurement of the Migdal effect in nuclear scattering. A low-pressure optical Time Projection Chamber is used to image in three-dimensions the characteristic signature of a Migdal event: an electron and a nuclear recoil track sharing a common vertex. Nuclear recoils are induced using fast neutrons from a D–D source, which scatter in the gaseous volume of the detector. The experiment is operated with 50 Torr of CF4 using two glass GEMs for charge amplification. Both light and charge are read-out, and these measurements are combined for track reconstruction. Commissioning data has been recorded with fast neutrons at the Neutron Irradiation Laboratory for Electronics (NILE) at Rutherford Appleton Laboratory in the UK. Results of the experiment’s commissioning and the performance of the detector with a high rate of highly ionising nuclear recoils are presented, along with results from low energy electrons. Initial results of light and charge read-out with low pressure noble gas mixtures are also presented.

CYGν\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ u $$\\end{document}S: detecting solar neutrinos with directional gas time projection chambers

Cygnus is a proposed global network of large-scale gas time projection chambers (TPCs) with the capability of directionally detecting nuclear and electron recoils at ≳\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\gtrsim $$\\end{document}keV energies. The primary focus of Cygnus so far has been the detection of dark matter, with directional sensitivity providing a means of circumventing the so-called “neutrino fog”. However, the excellent background rejection and electron/nuclear recoil discrimination provided by the 3-dimensional reconstruction of ionisation tracks could turn the solar neutrino background into an interesting signal in its own right. For example, directionality would facilitate the simultaneous spectroscopy of multiple different flux sources. Here, we evaluate the possibility of measuring solar neutrinos using the same network of gas TPCs built from 10 m3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^3$$\\end{document}-scale modules operating under conditions that enable simultaneous sensitivity to both dark matter and neutrinos. We focus in particular on electron recoils, which provide access to low-energy neutrino fluxes like pp, pep, 7\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^7$$\\end{document}Be, and CNO. An appreciable event rate is already detectable in experiments consisting of a single 10 m3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^3$$\\end{document} module, assuming standard fill gases such as CF4\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$_4$$\\end{document} mixed with helium at atmospheric pressure. With total volumes around 1000 m3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^3$$\\end{document} or higher, the TPC network could be complementary to dedicated neutrino observatories, whilst entering the dark-matter neutrino fog via the nuclear recoil channel. We evaluate the required directional performance and background conditions to observe, discriminate, and perform spectroscopy on neutrino events. We find that, under reasonable projections for planned technology that will enable 10–30-degree angular resolution and ∼10\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\sim 10$$\\end{document}% fractional energy resolution, Cygnus could be a competitive directional neutrino experiment.

Spatial resolution studies using point spread function extraction in optically read out Micromegas and GEM detectors

Optically read out gaseous detectors are used in track reconstruction and imaging applications requiring high granularity images. Among resolution-determining factors, the amplification stage plays a crucial role and optimizations of detector geometry are pursued to maximize spatial resolution. To compare Micro Pattern Gaseous Detector (MPGD) technologies, focused low-energy X-ray beams at the SOLEIL synchrotron facility were used to record and extract point spread function widths with MICRO-Mesh Gaseous Structure (icromegas) and Gas Electron Multiplier (GEM) detectors. Point spread function width of ≈108µm for Micromegas and ≈127µm for GEM foils were extracted. The scanning of the beam with different intensities, energies and across the detector active region can be used to quantify resolution-limiting factors and improve imaging detectors using MPGD amplification stages.

4D Track Reconstruction on Free-Streaming Data with PANDA at FAIR

A new generation of accelerator-based experiments in subatomic physics is being developed, where the challenge of separating rare signal processes from background at high intensities requires a change of trigger paradigm. At the future PANDA experiment at FAIR, hardware triggers will be abandoned and instead a purely software-based system will be used. This requires novel reconstruction methods with the ability to process data from many beam–target interactions simultaneously. A 4D tracking algorithm, based on the cellular automaton, has been developed which will utilize the timing information from detector signals. Simulation studies have been performed to test its performance on the foreseen free-streaming, i.e., continuous readout from the PANDA detector, where all processing is on the software level. For this purpose, a novel quality assurance method, tailored for tracking on free-streaming data has been developed and implemented. Benchmark studies on PANDA software show that at higher interaction rates, 4D tracking performs better than the 3D algorithm in terms of efficiency, 84%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} compared to 77%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}. In addition, the fake track suppression is greatly improved, reducing the rate of fake tracks of “ghosts” by about 50%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}.

Astronomy potential of KM3NeT/ARCA

The KM3NeT/ARCA neutrino detector is currently under construction at 3500 m depth offshore Capo Passero, Sicily, in the Mediterranean Sea. The main science objectives are the detection of high-energy cosmic neutrinos and the discovery of their sources. Simulations were conducted for the full KM3NeT/ARCA detector, instrumenting a volume of 1 km3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^3$$\\end{document}, to estimate the sensitivity and discovery potential to point-like neutrino sources. This paper covers the reconstruction of track- and shower-like signatures, as well as the criteria employed for neutrino event selection. With an angular resolution below 0.1∘\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^\\circ $$\\end{document} for tracks and under 2∘\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^\\circ $$\\end{document} for showers, the sensitivity to point-like neutrino sources surpasses existing observed limits across the entire sky.

Quantum-Annealing-Inspired Algorithms for Track Reconstruction at High-Energy Colliders

Charged particle reconstruction or track reconstruction is one of the most crucial components of pattern recognition in high-energy collider physics. It is known to entail enormous consumption of computing resources, especially when the particle multiplicity is high, which will be the conditions at future colliders, such as the High Luminosity Large Hadron Collider and Super Proton–Proton Collider. Track reconstruction can be formulated as a quadratic unconstrained binary optimization (QUBO) problem, for which various quantum algorithms have been investigated and evaluated with both a quantum simulator and hardware. Simulated bifurcation algorithms are a set of quantum-annealing-inspired algorithms, known to be serious competitors to other Ising machines. In this study, we show that simulated bifurcation algorithms can be employed to solve the particle tracking problem. The simulated bifurcation algorithms run on classical computers and are suitable for parallel processing and usage of graphical processing units, and they can handle significantly large amounts of data at high speed. These algorithms exhibit reconstruction efficiency and purity comparable to or sometimes improved over those of simulated annealing, but the running time can be reduced by as much as four orders of magnitude. These results suggest that QUBO models together with quantum-annealing-inspired algorithms are valuable for current and future particle tracking problems.

Detector-embedded reconstruction of complex primitives using FPGAs

The slowdown of Moore’s law and the growing requirements of future HEP experiments with ever-increasing data rates pose important computational challenges for data reconstruction and trigger systems, encouraging the exploration of new computing methodologies. In this work we discuss a FPGA-based tracking system, relying on a massively parallel pattern recognition approach, inspired by the processing of visual images by the natural brain (“retina architecture”). This method allows a large efficiency of utilisation of the hardware, low power consumption and very low latencies. Based on this approach, a device has been designed within the LHCb Upgrade-II project, with the goal of performing track reconstruction in the forward acceptance region in real-time during the upcoming Run 4 of the LHC. This innovative device will perform track reconstruction before the event-building, in a short enough time to provide pre-reconstructed tracks (“primitives”) transparently to the processor farm, as if they had been generated directly by the detector. This allows significant savings in higher-level computing resources, enabling handling higher luminosities than otherwise possible. The feasibility of the project is backed up by the results of tests performed on a realistic hardware prototype, that has been opportunistically processing actual LHCb data in parallel with the regular DAQ in the LHC Run 3.

2D and 3D analysis improvements with machine learning for muography applications

Transmission muography is a non-invasive and non-destructive imaging method which allows to estimate the integrated density of a volume in a given direction (also referred as opacity). It is used in multiple societal applications like archaeology, nuclear safety or geoscience. It relies on the reconstruction of muon tracks that crossed the studied volume compared to the corresponding open sky expectation. The portable experimental setups developed by CEA/Irfu group operates four Micromegas gaseous detectors, HV and DAQ modules, and an embedded computer allowing remote control.Used Micromegas detectors have a multiplexed readout to optimize the DAQ system, while keeping good spatial resolution. However, the natural muon flux is relatively low, so muography images could have high levels of statistical noise, which would be propagated to the reconstructed 3D images. For this reason, we propose three new methods, using machine learning, which increase significantly the quality of 2D images and 3D tomographies.We developed a new demultiplexing method for the Micromegas. It showed its efficiency both for 1D and 2D multiplexed detectors, in a hodoscopic tracker and a Time Projection Chamber (TPC). In the TPC, this method can embed a particle identification between muon-hits and electron-hits.We demonstrated how neural networks could denoise muography images. In the context of the scanning of a nuclear reactor, we used data augmentation, modeling few hundreds fake nuclear reactors. Muographies from different points of view were simulated to train a denoising neural network.These new methods significantly improved the muography images. Nonetheless, the used 3D tomography algorithm still has some limitations. We proved that it was possible to build a 3D post-process neural network, which was trained to compensate the reconstruction algorithm’s limitations.

The derivation of Jacobian matrices for the propagation of track parameter uncertainties in the presence of magnetic fields and detector material

In high-energy physics experiments, the trajectories of charged particles are reconstructed using track reconstruction algorithms. Such algorithms need to both identify the set of measurements from a single charged particle and to fit the parameters by propagating tracks along the measurements. The propagation of the track parameter uncertainties is an important component in the track fitting to get the optimal precision in the fitted parameters. The error propagation is performed at intersections between the track and local coordinate frames defined on detector components by calculating a Jacobian matrix corresponding to the local-to-local frame transport. This paper derives the Jacobian matrix in a general manner to harmonize with numerical integration methods developed for inhomogeneous magnetic fields and materials. The Jacobian and transported covariance matrices are validated by simulating the propagation of charged particles between two frames and comparing with the results of numerical methods.

Attentional modulation of secondary somatosensory and visual thalamus of mice.

Each sensory modality has its own primary and secondary thalamic nuclei. While the primary thalamic nuclei are well understood to relay sensory information from the periphery to the cortex, the role of secondary sensory nuclei is elusive. One hypothesis has been that secondary nuclei may support feature-based attention. If this is true, one would also expect the activity in different nuclei to reflect the degree to which modalities are or are not behaviorally relevant in a task. We trained head-fixed mice to attend to one sensory modality while ignoring a second modality, namely to attend to touch and ignore vision, or vice versa. Arrays were used to record simultaneously from secondary somatosensory thalamus (POm) and secondary visual thalamus (LP). In mice trained to respond to tactile stimuli and ignore visual stimuli, POm was robustly activated by touch and largely unresponsive to visual stimuli. A different pattern was observed when mice were trained to respond to visual stimuli and ignore touch, with POm now more robustly activated during visual trials. This POm activity was not explained by differences in movements (i.e., whisking, licking, pupil dilation) resulting from the two tasks. Post hoc histological reconstruction of array tracks through POm revealed that subregions varied in their degree of plasticity. LP exhibited similar phenomena. We conclude that behavioral training reshapes activity in secondary thalamic nuclei. Secondary nuclei may respond to behaviorally relevant, reward-predicting stimuli regardless of stimulus modality.

Development of a Vector Finder Toolkit for Track Reconstruction at NICA Experiments: Current Status and Future Prospects

Development of a Vector Finder Toolkit for Track Reconstruction at NICA Experiments: Current Status and Future Prospects

Simulation and reconstruction of particle trajectories in the CEPC drift chamber

The circular electron-positron collider (CEPC) is designed to precisely measure the properties of the Higgs boson, study electroweak interactions at the Z-boson peak, and search for new physics beyond the Standard Model. As a component of the 4th\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$4^{\ ext {th}}$$\\end{document} conceptual CEPC detector, the drift chamber facilitates the measurement of charged particles. This study implemented a Geant4-based simulation and track reconstruction for the drift chamber. For the simulation, detector construction and response were implemented and added to the CEPC simulation chain. The development of track reconstruction involves track finding using the combinatorial Kalman filter method and track fitting using the tool of GenFit. Using the simulated data, the tracking performance was studied. The results showed that both the reconstruction resolution and tracking efficiency satisfied the requirements of the CEPC experiment.

Characteristics of the prototype LAMPS TPC for low-energy nuclear collisions at RAON

The time projection chambers (TPCs) with triple GEM amplifiers are tested at HIMAC using 12C beams to investigate their nucleus track reconstruction capability. The two identical TPCs, each having an active volume of 96 × 64 × 144 mm3 and rectangular pixel pads, have been developed as prototypes of the LAMPS collaboration’s low-energy active target TPC (AT-TPC). The responses to light nuclei, such as particle identification and spatial resolution, have been investigated. The Si-CsI detectors located next to the TPCs further facilitate the identification of incident particles, as well as the matching tracks. This paper presents the technical details of the experimental setup, pulse shape analysis, tracking algorithm, and spatial resolution measurement. Comparisons of the experimental data with the calculations by the Monte Carlo (MC) simulation are also discussed.

Tracking the porphyry-epithermal mineralization transition using U-Pb carbonate dating

Key metals important for the green energy transition concentrate during magmatic-hydrothermal processes. In porphyry deposits, epithermal mineralization can overprint earlier higher-temperature systems. It is not well understood whether mineralization occurs in a single evolving system or forms during pulsed, episodic overprinting events. The timing and duration of fluid flow therefore remain key data gaps in deposit models, but they are essential factors for understanding metal (re)mobilization and concentration processes. Carbonates are common gangue minerals that precipitate during fluid flow and can be dated using the U-Pb method, thereby directly dating hydrothermal processes. Here, 41 new U-Pb dates from a fault-controlled porphyry-epithermal system in Yukon, Canada, reveal a >50 m.y. record of carbonate precipitation between ca. 77 and 19 Ma. Results support a model of pulsed, episodic fluid flow, rather than a single evolving system, where epithermal carbonate precipitation at ca. 74−67 Ma was both coeval with and significantly postdated Cretaceous porphyry-related magmatism. Overprinting events at ca. 62−56 Ma, ca. 51−47 Ma, and younger than 40 Ma were not responsible for primary metal deposition but may have contributed to metal enrichment. Carbonate dates coincide with periods of brecciation and fault slip. Fault movement therefore enabled episodic overprinting by epithermal mineralization, mobilizing and (re)concentrating metals. This comprehensive reconstruction of a long-lived magmatic-hydrothermal system tracks the transition from porphyry to epithermal environments, demonstrating the power of carbonate U-Pb dating for critical minerals research.

SGRTmreg: A Learning-Based Optimization Framework for Multiple Pairwise Registrations.

Point cloud registration is a fundamental task in computer vision and graphics, which is widely used in 3D reconstruction, object tracking, and atlas reconstruction. Learning-based optimization and deep learning methods have been widely developed in pairwise registration due to their own distinctive advantages. Deep learning methods offer greater flexibility and enable registering unseen point clouds that are not trained. Learning-based optimization methods exhibit enhanced robustness and stability when handling registration under various perturbations, such as noise, outliers, and occlusions. To leverage the strengths of both approaches to achieve a less time-consuming, robust, and stable registration for multiple instances, we propose a novel computational framework called SGRTmreg for multiple pairwise registrations in this paper. The SGRTmreg framework utilizes three components-a Searching scheme, a learning-based optimization method called Graph-based Reweighted discriminative optimization (GRDO), and a Transfer module to achieve multi-instance point cloud registration.Given a collection of instances to be matched, a template as a target point cloud, and an instance as a source point cloud, the searching scheme selects one point cloud from the collection that closely resembles the source. GRDO then learns a sequence of regressors by aligning the source to the target, while the transfer module stores and applies the learned regressors to align the selected point cloud to the target and estimate the transformation of the selected point cloud. In short, SGRTmreg harnesses a shared sequence of regressors to register multiple point clouds to a target point cloud. We conduct extensive registration experiments on various datasets to evaluate the proposed framework. The experimental results demonstrate that SGRTmreg achieves multiple pairwise registrations with higher accuracy, robustness, and stability than the state-of-the-art deep learning and traditional registration methods.