cs.ROOct 30, 2026

A Capability-Aware Backup-Recoverability Filter for Heterogeneous Multi-Robot Navigation

Authors: Joonkyung Kim, Yanze Zhang, Wenhao Luo, Yiwei Lyu

Organizations: Department of Computer Science and Engineering, Texas A&M University · Department of Computer Science, University of Illinois Chicago

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Safe navigation in heterogeneous multi-robot teams faces two coupled challenges: differences in robots' braking and turning capabilities limit how each robot can respond to an approaching neighbor, while simultaneous interactions require a single control action to satisfy multiple collision-avoidance constraints. Together, these challenges can make local safety constraints incompatible with a robot's admissible controls, leaving a decentralized safety controller without a feasible action. We propose a Capability-Aware Backup-Recoverability Filter (CA-BRF), based on the insight that navigation actions should preserve access to a certified recovery maneuver. First, we construct a backup-recoverable set in a shared Cartesian operational space by evaluating whether prescribed backup maneuvers maintain pairwise safety until the robots come to rest. Each maneuver respects the corresponding robot's kinematic and actuation constraints, making the recovery certificate capability-dependent. Second, CA-BRF filters candidate controls by selecting admissible interpolations between candidate and backup actions while coordinating the selection across simultaneous interactions. This preserves backup recoverability while retaining as much of the candidate action as possible and ensures that all pairwise recovery conditions are satisfied by a single executable control for each robot. Together, we establish recursive feasibility and safety for the discrete-time closed-loop system even when the nominal safety-constrained optimization becomes infeasible. The validity and effectiveness of the proposed CA-BRF are validated in simulations with up to 30 heterogeneous robots and experiments of robots with five different kinematic classes.

Figures & tables

Figure 1: Real-robot trajectory comparison, recovery-certificate margin, and model-dependent velocity sets. Five heterogeneous robots (DI, UNI, DD, CL, and FO) navigate to individual goals. (a) With CA-BRF (ours), the robots navigate successfully while maintaining the prescribed pairwise safety condition. (b) Directly applying the HOCBF candidate controls without CA-BRF violates this condition. Darker trajectory colors indicate later times. (c) Backup safety margin Bijψ(x)B_{ij}^{\psi}(\mathbf{x}) for the CL–FO pair under the candidate and CA-BRF applied controls. Values below zero indicate that the prescribed backup maneuvers no longer certify safe separation. The candidate control crosses below zero, whereas the CA-BRF applied control remains nonnegative, preserving backup recoverability. The shaded interval marks the active CL–FO interaction during which CA-BRF modifies the candidate control. (d) Feasible (v,ω)(v,\omega) velocity sets and constraints for four representative nonholonomic kinematic classes [ 1 ] ; DI is omitted as holonomic.
Table 2
Figure 2: Random start–goal navigation with N=30N=30 heterogeneous robots (six per model) at multiple timestamps. Robots start from randomly sampled positions, and ×\times markers denote their assigned goals. UNI, DI, DD, CL, and FO robots are shown as blue circles, light-blue circles, green diamonds, red rectangles, and purple sectors, respectively. Dashed outlines show the operational-space safety radii rir_{i} centered at the corresponding operational points (Sec. II ).
MetricN=20N=20N=30N=30
Decision differs (%)4.7411.61
Shared accepts / CA-BRF rejects (%)4.5811.25
Shared rejects / CA-BRF accepts (%)0.210.48
Table III : Same-state comparison of model-specific and shared recovery evaluation. All controller components are fixed except for the backup rollout used to evaluate recoverability.
 N=20N=20N=30N=30
CertificateRecall (%)Lead (s)Recall (%)Lead (s)
BhB^{h} only76.90.1567.00.10
BhB^{h} and BψB^{\psi}100.00.8597.80.85
Table IV : Early-warning ability of the recovery certificate on HOCBF-only trajectories. Recall is the fraction of safety-buffer breaches preceded by a certificate violation within the previous 4040 control steps ( 2.02.0 s), and lead time is the median time from certificate violation to breach.
Figure 3: Effect of coordinating filter decisions across simultaneously interacting robots. CA-BRF uses one common candidate–backup interpolation factor within each connected interaction component, whereas the comparison variant selects factors independently. (a) Fraction of trials with composition failures, in which locally accepted decisions fail the pairwise backup-recoverability test when executed together. (b) Fraction of active robot control steps applying the full backup ( λ=0\lambda=0 ). Coordinated selection removes the observed composition failures while requiring more frequent full-backup activation in dense interactions.
More in this topic

Multi-Robot Navigation in Unknown Environments

See all 15 papers

Explore similar work

CardsList

  1. Capability-Aware Heterogeneous Control Barrier Functions for Decentralized Multi-Robot Safe Navigation

    Apr 14, 2026Joonkyung Kim, Yanze Zhang, Wenhao Luo +1Safe Robot NavigationControl Barrier Functions

    Department of Computer Science and Engineering, Texas A&M University · Department of Computer Science, University of Illinois Chicago

  2. Exact Feasibility Certification and Optimal Responsibility Allocation for Multi-Robot CBF Safety Filters

    Sep 14, 2026Chandan Kumar Sah, Jishnu KeshavanSafety-Critical ControlControl Barrier Functions

    Department of Mechanical Engineering, Indian Institute of Science, Bengaluru 560012, India

  3. Capability-Conditioned Graph Control Barrier Functions for Zero-Shot Safe Navigation of Heterogeneous Mobile Robot Teams

    Oct 30, 2026Joonkyung Kim, Janghyun Cho, Jeehyun Lee +2Multi-Robot Navigation in Unknown EnvironmentsMulti-Robot Systems

    Dept. of Computer Science and Engineering, Texas A&M University. · Dept. of Artificial Intelligence, Sogang University. · Dept. of Electronic Engineering, Sogang University.