Robust Global-Local Behavior Arbitration via Continuous Command Fusion Under LiDAR Errors

Mohamed Elgouhary; Amr S. El-Wakeel

cs.ROMar 28, 2026

Robust Global-Local Behavior Arbitration via Continuous Command Fusion Under LiDAR Errors

Authors: Mohamed Elgouhary, Amr S. El-Wakeel

Organizations: The Authors are with the Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV, USA.

Abstract

Modular autonomous driving systems must coordinate global progress objectives with local safety-driven reactions under imperfect sensing and strict real-time constraints. This paper presents a ROS2-native arbitration module that continuously fuses the outputs of two unchanged and interpretable controllers: a global reference-tracking controller based on Pure Pursuit and a reactive LiDAR-based Gap Follow controller. At each control step, both controllers propose Ackermann commands, and a PPO-trained policy predicts a continuous gate from a compact feature observation to produce a single fused drive command, augmented with practical safety checks. For comparison under identical ROS topic inputs and control rate, we implement a lightweight sampling-based predictive baseline. Robustness is evaluated using a ROS2 impairment protocol that injects LiDAR noise, delay, and dropout, and additionally sweeps forward-cone false short-range outliers. In a repeatable close-proximity passing scenario, we report safe success and failure rates together with per-step end-to-end controller runtime as sensing stress increases. The study is intended as a command-level robustness evaluation in a modular ROS2 setting, not as a replacement for planning-level interaction reasoning.

Figures & tables

Node	Subscribes	Publishes
Pure Pursuit (ego)	/ego_racecar/odom	/pure_pursuit_cmd
Gap Follow (ego)	/scan	/gap_follow_cmd
Behavior Arbiter	/scan (or /scan_imp ) /ego_racecar/odom /opp_racecar/odom /pure_pursuit_cmd /gap_follow_cmd	/drive
Sampling MPC	/scan /ego_racecar/odom	/drive /mpc_ref_marker /mpc_pred_marker /mpc_path_marker
Sensor Impairment	/scan /opp_racecar/odom	/scan_imp /opp_racecar/ odom_imp
RaceEval (metrics)	/scan /ego_racecar/odom /opp_racecar/odom /drive	JSON logs; summary metrics

TABLE I: ROS2 stack and topic interfaces used in simulation experiments.

Fig. 1: System overview. Pure Pursuit (PP) provides global reference tracking, while Gap Follow (GF) provides local LiDAR-based reactive collision avoidance. A PPO-trained arbiter outputs a continuous gate $\alpha\in[0,1]$ that fuses their candidate Ackermann commands into a single drive command. A safety monitor can override the fused command when clearance or data-health checks fail.

Parameter	Value
Total steps	$1.2\times 10^{6}$
Rollout length $n_{\mathrm{steps}}$	$4096$
Batch size $B$	$256$
Epochs per update $n_{\mathrm{epochs}}$	$5$
Discount $\gamma$	$0.99$
GAE $\lambda$	$0.98$

TABLE II: PPO training hyperparameters.

Setting	Value
Horizon steps $N$	$8$
Timestep $\Delta t$	$0.1$ s
Candidates $M$ (standard tracking)	$9$
Candidates $M$ (interaction)	$17$
Steering limits $\delta$	$[-0.4189,\,0.4189]$ rad
Safety radius $r_{\text{safe}}$	$0.55$ m

TABLE III: Sampling-based predictive baseline parameters.

Item	Setting
Simulator	F1TENTH Gym ROS (ROS2)
Tracks	Train: Levine ; Eval: unseen IMS
Control rate	$f_{c}=30$ Hz ( $T_{s}=33.3$ ms)
Base LiDAR range noise	$\sigma=0.05$ m (zero-mean Gaussian; clipped to $[r_{\min},r_{\max}]$ )
Base LiDAR scan delay	$\tau_{\text{scan}}=200$ ms (FIFO queue)
Base LiDAR dropout	$p_{\text{drop}}=0.3$ per scan (zero-order hold / last-scan republish)

TABLE IV: Experimental design: LiDAR impairment parameters, repeated trials, and runtime measurement setup.

Fig. 2: Qualitative interaction sequence in simulation. The ego vehicle transitions from global reference tracking to a local free-space maneuver to pass the leading vehicle, then returns to reference tracking.

Method	SR (%)	Coll. (%)	OffTrk (%)	T/O (%)	Mean runtime (ms)	Worst runtime (ms)
Arbiter (PP+GF)	100.0	0.0	0.0	0.0	31.95	47.12
Sampling MPC	90.0	0.0	0.0	10.0	24.15	61.83

TABLE V: Baseline interaction performance on IMS in F1TENTH Gym ROS (

N_{\text{heats}}=10

Fig. 3: Robustness under base LiDAR impairments (noise, 200 ms delay, 0.3 dropout) with front-cone outliers swept over $p_{\text{out}}\in\{0,0.2,0.4\}$ .

Autonomous Driving

Autonomous driving develops vehicles that perceive road scenes, predict interactions, and plan and control safe movement with limited or no human input.

See all 536 papers

DeepIPCv3: Event-Aware Multi-Modal Sensor Fusion for Sudden Pedestrian Crossing Avoidance
May 31, 2026Oskar Natan, Andi Dharmawan, Aufaclav Zatu Kusuma Frisky +2Collision Avoidance Camera-LiDAR Fusion
Department of Computer Science and Electronics, Universitas Gadjah Mada, Yogyakarta 55281, Indonesia
Simplifying ROS2 controllers with a modular architecture for robot-agnostic reference generation
Jan 13, 2026Davide Risi, Vincenzo Petrone, Antonio Langella +3Robotic Control Robot Trajectory Tracking
Department of Information Engineering, Electrical Engineering and Applied Mathematics (DIEM), University of Salerno, 84084 Fisciano, Italy
A Numerically-Robust ROS 2 Port of iG-LIO: Diagnosing and Fixing Toolchain-Induced Failures in Incremental GICP LiDAR-Inertial Odometry
Jul 10, 2026Afonso E. Carvalho, David Portugal, Paulo PeixotoLiDAR Odometry Robotics
Robotics Institute (RI), Carnegie Mellon University (CMU), Pittsburgh, USA. · Institute of Systems and Robotics (ISR), University of Coimbra (UC), Coimbra, Portugal.

May 31, 2026J/K move · Enter open · S save

DeepIPCv3: Event-Aware Multi-Modal Sensor Fusion for Sudden Pedestrian Crossing Avoidance

Oskar Natan, Andi Dharmawan, Aufaclav Zatu Kusuma Frisky, Jazi Eko Istiyanto+1

Department of Computer Science and Electronics, Universitas Gadjah Mada, Yogyakarta 55281, Indonesia

Collision Avoidance Camera-LiDAR Fusion Autonomous Driving Planning Dynamic Obstacle Avoidance

Open paper AskPDF arXiv

Current end-to-end autonomous driving systems predominantly rely on frame-based sensors, which suffer from inherent perception latency and motion blur during highly dynamic encounters, specifically sudden pedestrian crossings. To address this critical safety vulnerability, we propose DeepIPCv3, a novel multi-modal autonomous navigation framework that synergizes the dense 3D spatial geometry of LiDAR point clouds with the microsecond-level asynchronous event streams of a Dynamic Vision Sensor (DVS). We introduce a Transformer-inspired cross-modal attention mechanism to dynamically correlate these distinct modalities, allowing the network to instantaneously prioritize high-speed dynamic updates without sacrificing structural scene awareness. The fused latent representations are then mapped to safe local waypoints and executable control commands via a hybrid policy network that blends heuristic trajectory tracking with direct neural predictions. Due to the severe physical risks associated with live testing of these sudden crossing scenarios, the framework is rigorously evaluated offline using a custom multi-modal dataset collected across both well-illuminated noon and challenging evening conditions. Extensive comparative and ablation studies demonstrate that DeepIPCv3 achieves state-of-the-art predictive performance. By effectively eliminating exposure failures and motion blur, the proposed LiDAR and DVS fusion yields the lowest trajectory and control command errors, enabling highly reactive, mathematically bounded evasive maneuvers regardless of ambient illumination. To support future research, we will release the codes to our GitHub repo at https://github.com/oskarnatan/DeepIPCv3.

Abstract

Autonomous Driving

Explore similar work

DeepIPCv3: Event-Aware Multi-Modal Sensor Fusion for Sudden Pedestrian Crossing Avoidance

Simplifying ROS2 controllers with a modular architecture for robot-agnostic reference generation

A Numerically-Robust ROS 2 Port of iG-LIO: Diagnosing and Fixing Toolchain-Induced Failures in Incremental GICP LiDAR-Inertial Odometry