COHORT: Hybrid RL for Collaborative Large DNN Inference on Multi-Robot Systems Under Real-Time Constraints

• URL: http://arxiv.org/abs/2603.10436v1
• Date: Wed, 11 Mar 2026 05:38:00 GMT
• Title: COHORT: Hybrid RL for Collaborative Large DNN Inference on Multi-Robot Systems Under Real-Time Constraints
• Abstract summary: Large deep neural networks (DNNs) are computationally demanding and challenging to deploy on resource-constrained edge platforms like field robots.<n>We present COHORT, a collaborative DNN inference and task-execution framework for multi-robot systems built on the Robotic Operating System (ROS)<n> Experimental results demonstrate that COHORT reduces battery consumption by 15.4% and increases GPU utilization by 51.67%, while satisfying frame-rate and deadline constraints 2.55 times of the time.
• Score: 0.769672852567215
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large deep neural networks (DNNs), especially transformer-based and multimodal architectures, are computationally demanding and challenging to deploy on resource-constrained edge platforms like field robots. These challenges intensify in mission-critical scenarios (e.g., disaster response), where robots must collaborate under tight constraints on bandwidth, latency, and battery life, often without infrastructure or server support. To address these limitations, we present COHORT, a collaborative DNN inference and task-execution framework for multi-robot systems built on the Robotic Operating System (ROS). COHORT employs a hybrid offline-online reinforcement learning (RL) strategy to dynamically schedule and distribute DNN module execution across robots. Our key contributions are threefold: (a) Offline RL policy learning combined with Advantage-Weighted Regression (AWR), trained on auction-based task allocation data from heterogeneous DNN workloads across distributed robots, (b) Online policy adaptation via Multi-Agent PPO (MAPPO), initialized from the offline policy and fine-tuned in real time, and (c) comprehensive evaluation of COHORT on vision-language model (VLM) inference tasks such as CLIP and SAM, analyzing scalability with increasing robot/workload and robustness under . We benchmark COHORT against genetic algorithms and multiple RL baselines. Experimental results demonstrate that COHORT reduces battery consumption by 15.4% and increases GPU utilization by 51.67%, while satisfying frame-rate and deadline constraints 2.55 times of the time.

Related papers

• $τ_0$-VLA: a Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Computation [51.05516457377339]
  Long-horizon robot manipulation requires a robot to both execute individual skills reliably and sequence them coherently over extended tasks.<n>We introduce $_0$-VLA, a hierarchical robot foundation model that formulates high-level subtask generation as a compute-scalable inference problem.<n>A low-level policy then executes the generated subtask across multiple robot embodiments.
  arXiv  Detail & Related papers  (2026-08-17T17:59:11Z)
• Population-Coded Spiking Neural Networks for High-Dimensional Robotic Control [0.9916299448089259]
  This paper introduces a novel framework combining population-coded Spiking Neural Networks (SNNs) with Deep Reinforcement Learning (DRL)<n>We show that our approach achieves energy savings of up to 96.10% compared to traditional Artificial Neural Networks (ANNs) while maintaining comparable control performance.<n>These results position population-coded SNNs as a promising solution for energy-efficient, high-performance robotic control in resource-constrained applications.
  arXiv  Detail & Related papers  (2025-10-12T09:27:25Z)
• Exploring the Dynamic Scheduling Space of Real-Time Generative AI Applications on Emerging Heterogeneous Systems [0.9041154551329587]
  Real-time generative AI (RTGen) workloads combine the compute intensity and dynamic execution patterns of generative models with the stringent latency and constraints of real-time inference.<n>Modern edge platforms increasingly adopt heterogeneous system-on-chip (SoC) architectures.<n>We show that scheduling decisions significantly affect workload performance.
  arXiv  Detail & Related papers  (2025-07-19T18:24:11Z)
• MAGNNET: Multi-Agent Graph Neural Network-based Efficient Task Allocation for Autonomous Vehicles with Deep Reinforcement Learning [2.5022287664959446]
  We introduce a novel framework that integrates graph neural networks (GNNs) with a centralized training and decentralized execution (CTDE) paradigm.<n>Our approach enables unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) to dynamically allocate tasks efficiently without necessitating central coordination.
  arXiv  Detail & Related papers  (2025-02-04T13:29:56Z)
• FusionLLM: A Decentralized LLM Training System on Geo-distributed GPUs with Adaptive Compression [55.992528247880685]
  Decentralized training faces significant challenges regarding system design and efficiency. We present FusionLLM, a decentralized training system designed and implemented for training large deep neural networks (DNNs) We show that our system and method can achieve 1.45 - 9.39x speedup compared to baseline methods while ensuring convergence.
  arXiv  Detail & Related papers  (2024-10-16T16:13:19Z)
• DNN Partitioning, Task Offloading, and Resource Allocation in Dynamic Vehicular Networks: A Lyapunov-Guided Diffusion-Based Reinforcement Learning Approach [49.56404236394601]
  We formulate the problem of joint DNN partitioning, task offloading, and resource allocation in Vehicular Edge Computing. Our objective is to minimize the DNN-based task completion time while guaranteeing the system stability over time. We propose a Multi-Agent Diffusion-based Deep Reinforcement Learning (MAD2RL) algorithm, incorporating the innovative use of diffusion models.
  arXiv  Detail & Related papers  (2024-06-11T06:31:03Z)
• SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning [82.46975428739329]
  We develop a library containing a sample efficient off-policy deep RL method, together with methods for computing rewards and resetting the environment.<n>We find that our implementation can achieve very efficient learning, acquiring policies for PCB board assembly, cable routing, and object relocation.<n>These policies achieve perfect or near-perfect success rates, extreme robustness even under perturbations, and exhibit emergent robustness recovery and correction behaviors.
  arXiv  Detail & Related papers  (2024-01-29T10:01:10Z)
• Intelligent Trajectory Design for RIS-NOMA aided Multi-robot Communications [59.34642007625687]
  The goal is to maximize the sum-rate of whole trajectories for multi-robot system by jointly optimizing trajectories and NOMA decoding orders of robots. An integrated machine learning (ML) scheme is proposed, which combines long short-term memory (LSTM)-autoregressive integrated moving average (ARIMA) model and dueling double deep Q-network (D$3$QN) algorithm.
  arXiv  Detail & Related papers  (2022-05-03T17:14:47Z)
• Semantic-Aware Collaborative Deep Reinforcement Learning Over Wireless Cellular Networks [82.02891936174221]
  Collaborative deep reinforcement learning (CDRL) algorithms in which multiple agents can coordinate over a wireless network is a promising approach. In this paper, a novel semantic-aware CDRL method is proposed to enable a group of untrained agents with semantically-linked DRL tasks to collaborate efficiently across a resource-constrained wireless cellular network.
  arXiv  Detail & Related papers  (2021-11-23T18:24:47Z)
• RAPID-RL: A Reconfigurable Architecture with Preemptive-Exits for Efficient Deep-Reinforcement Learning [7.990007201671364]
  We propose a reconfigurable architecture with preemptive exits for efficient deep RL (RAPID-RL) RAPID-RL enables conditional activation of preemptive layers based on the difficulty level of inputs. We show that RAPID-RL incurs 0.34x (0.25x) number of operations (OPS) while maintaining performance above 0.88x (0.91x) on Atari (Drone navigation) tasks.
  arXiv  Detail & Related papers  (2021-09-16T21:30:40Z)
• Deep Reinforcement Learning with Population-Coded Spiking Neural Network for Continuous Control [0.0]
  We propose a population-coded spiking actor network (PopSAN) trained in conjunction with a deep critic network using deep reinforcement learning (DRL) We deployed the trained PopSAN on Intel's Loihi neuromorphic chip and benchmarked our method against the mainstream DRL algorithms for continuous control. Our results support the efficiency of neuromorphic controllers and suggest our hybrid RL as an alternative to deep learning, when both energy-efficiency and robustness are important.
  arXiv  Detail & Related papers  (2020-10-19T16:20:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.