Collective dynamics of active matter with orientation-weighted alignment

Bohdan Dobosh, Alexander Yakimenko · 2026

Read Collective dynamics of active matter with orientation-weighted alignment by Bohdan Dobosh, Alexander Yakimenko (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-17627.txt.

Introduction

Dobosh and Yakimenko study an agent-based simulation of self-propelled particles with a speed-dependent alignment rule. The interaction is angle-weighted and acts along the line joining neighboring particles. Tuning its strength yields distinct collective regimes: disordered gas-like motion, coherent flock movement, jammed high-density phases, and densely ordered moving sets with active-crystal-like behavior. A simple local rule thus generates a broad range of group dynamics within one tiny model.

Catalog record

CREATORBohdan Dobosh, Alexander Yakimenko
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE35374
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.17627

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Collective dynamics of active matter with orientation-weighted alignment Bohdan Dobosh, Alexander Yakimenko Abstract We study an agent-based model of self-propelled particles with a velocity-dependent alignment rule. This interaction is orientation weighted and acts along the line connecting neighboring particles. Tuning the alignment strength produces several distinct collective regimes, including disordered gas-like motion, coherent flocking, jammed high-density states, and densely ordered moving clusters with active-crystal-like behavior. These results show that a simple local alignment rule can generate a broad range of nonequilibrium collective dynamics within a single microscopic model. I Introduction Active matter is a class of nonequilibrium systems composed of self-propelled particles or agents that continuously convert internal or environmental energy into directed motion and thus generate collective behavior far from thermal equilibrium 29 ; 16 ; 21 . Examples range from bacterial suspensions, cell layers, and animal groups to synthetic systems such as colloidal rollers, magnetic colloids, vibrated granular media, and active nematics 32 ; 6 ; 5 ; 19 ; 4 ; 26 ; 27 ; 22 . Simple local interactions in such systems can produce robust large-scale collective behavior, including flocking, clustering, jamming, and dense ordered motion 36 ; 34 ; 33 ; 2 . Local alignment is one of the main mechanisms underlying collective motion in active matter. The Vicsek model and its extensions showed that short-range alignment can generate polar order and nonequilibrium transitions in systems of self-propelled particles 36 ; 34 . When alignment is combined with attractive or repulsive interactions, the collective behavior becomes richer, leading to cohesive flocks, clustered states, and dense moving aggregates 17 ; 10 . Related phenomena have also been observed in synthetic active matter. Colloidal rollers display flocking and vortex formation 6 ; 5 , while active colloids can form long-lived dense structures, including living crystals and other stable collective states 28 ; 25 ; 3 . At the continuum and mesoscopic levels, dense active motion has been described in terms of active crystals, active solids, and dense active-matter models of collective motion 15 ; 23 ; 24 ; 18 , and two-dimensional active systems are now known to exhibit a broad range of liquid, hexatic, solid, and phase-separated regimes 20 ; 8 ; 13 ; 12 . These developments motivate the search for simple microscopic interaction rules that can generate different collective regimes within a single dynamical framework. Most Vicsek-type alignment models prescribe alignment through angular averaging of neighboring headings, although many extensions modify the interaction symmetry, include cohesion, or add additional force-based ingredients 36 ; 11 ; 33 ; 2 . Here we instead introduce a force-based alignment rule: the contribution from a neighbor acts along the line connecting the two particles, while its strength and sign are determined by the projection of the neighbor’s velocity onto this line. Since this projection depends on both the neighbor’s heading and speed, the response is controlled not only by how the neighbor moves, but also by where it is located relative to the particle under consideration 33 ; 2 ; 37 . This rule differs from heading-averaging alignment, and also from many force-based cohesive or selective-response models in which attraction–repulsion, escape–pursuit, and alignment mechanisms are introduced as separate prescribed interactions 17 ; 30 ; 31 ; 33 ; 2 . In the present model, the effective alignment force is tied directly to the velocity projection on the interparticle axis. The same neighbor can therefore generate an acceleration either toward or away from it, and can either increase or decrease the particle speed, depending on the relative position and direction of motion. This geometry-dependent coupling introduces a front–back asymmetry, absent in the minimal Vicsek alignment rule, and directly couples spatial organization, acceleration, and alignment. The model therefore provides a compact setting for studying how anisotropic velocity-dependent interactions can produce flocking, dispersive motion, dense ordering, and collective turning within the same microscopic description. Such anisotropic response is also relevant to biological collectives. Experiments on bird flocks and insect groups have shown that directional changes and collective responses to perturbations can propagate through a group via local interactions and behavioral inertia, without requiring designated leaders 1 ; 14 ; 35 ; 38 . More recent work has further emphasized that the response to neighbors can depend on relative position and directional perception, rather than on heading alone 37 ; 7 . This motivates studying models in which coherent reversals arise from the interaction rule itself and from the response of the group to localized external forcing. In this work we introduce an agent-based model of self-propelled particles with three ingredients: orientation-weighted alignment, short-range Lennard-Jones interactions, and speed saturation. In the absence of external forcing, the sign of the alignment strength separates coherent flocking from dispersive gas-like motion. We then use an anisotropic target field, constructed in analogy with the same orientation-weighted interaction, to probe how an ordered group responds to spatially localized cues. Depending on the target strength and alignment parameters, the system evolves toward either jammed high-density states or densely ordered moving clusters with local sixfold structure and coherent reversals. The model therefore provides a compact framework for studying how geometry-dependent alignment, short-range interactions, and anisotropic forcing shape nonequilibrium collective dynamics. The paper is organized as follows. Section II introduces the model and simulation procedure. Section III presents the main collective regimes and discusses stochasticity and effective friction. Section IV summarizes the results. II Model We consider a system of N self-propelled agents moving in two spatial dimensions. Time is discretized with step \Delta t , while particle positions remain continuous. The dynamics are described by \ddot{\mathbf{r}}_{i}=\mathbf{a}_{i}^{\mathrm{int}}+\mathbf{a}_{i}^{\mathrm{bord}}+\mathbf{a}_{i}^{\mathrm{fl}}, (1) where \mathbf{r}_{i} , \mathbf{v}_{i} , and \mathbf{a}_{i} denote the position, velocity, and acceleration of the i -th agent. The three terms on the right-hand side represent interparticle interactions, confinement by the boundary, and alignment-driven motion, respectively. Figure 1: (a) Lennard-Jones-type pair potential associated with a central agent (green triangle). The red-shaded region corresponds to short-range repulsion that prevents collisions, while the green-shaded region indicates the attractive part of the interaction. (b) Alignment force field generated by a single agent located at the center (green triangle). Arrows show the force direction, and the background color represents the absolute value of the force. Panels (c) and (d) illustrate the dynamics of a two-agent system with the same initial conditions, shown by empty red and blue triangles, but opposite signs of the alignment parameter A ; the corresponding trajectories are shown by solid red and dashed blue lines. (c) For A>0 , the agents become aligned and move with a common velocity. (d) For A<0 , the agents first become anti-aligned and eventually repel each other. Pair interactions are described by a truncated Lennard-Jones-type potential, shown in Fig. 1 (a). The parameter r_{\rm cut} defines the interaction range: only particles with r_{ij}=|\mathbf{r}_{j}-\mathbf{r}_{i}|\leq r_{\rm cut} contribute to the pair potential and alignment force. The interaction is repulsive at short distances, attractive at intermediate distances, and has a minimum at the equilibrium separation r_{0} . For r_{ij}\leq r_{\mathrm{cut}} , we write U_{LJ}(r_{ij})=U_{p}\left[\left(\frac{r_{0}}{r_{ij}}\right)^{12}-2\left(\frac{r_{0}}{r_{ij}}\right)^{6}\right]+U_{\mathrm{shift}}, (2) while U_{LJ}(r_{ij})=0 for r_{ij}>r_{\mathrm{cut}} . Here U_{p} sets the interaction strength, and the constant shift U_{\mathrm{shift}} is chosen so that U_{LJ}(r_{\mathrm{cut}})=0 . The corresponding interaction acceleration acting on the i -th agent is \mathbf{a}_{i}^{\mathrm{int}}=-\nabla_{i}\sum_{j\neq

Read the complete text