Can SSD-Mamba2 Unlock Reinforcement Learning for End-to-End Motion Control?

• URL: http://arxiv.org/abs/2509.07593v1
• Date: Tue, 09 Sep 2025 11:05:44 GMT
• Title: Can SSD-Mamba2 Unlock Reinforcement Learning for End-to-End Motion Control?
• Abstract summary: We present a vision-driven cross-modal framework built on SSD-Mamba2, a selective state-space backbone.<n>Proprioceptive states and exteroceptive observations are encoded into compact tokens and fused by stacked SSD-Mamba2 layers.<n>Our approach consistently surpasses strong state-of-the-art baselines in return, safety, and sample efficiency.
• Score: 2.211616153144189
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: End-to-end reinforcement learning for motion control promises unified perception-action policies that scale across embodiments and tasks, yet most deployed controllers are either blind (proprioception-only) or rely on fusion backbones with unfavorable compute-memory trade-offs. Recurrent controllers struggle with long-horizon credit assignment, and Transformer-based fusion incurs quadratic cost in token length, limiting temporal and spatial context. We present a vision-driven cross-modal RL framework built on SSD-Mamba2, a selective state-space backbone that applies state-space duality (SSD) to enable both recurrent and convolutional scanning with hardware-aware streaming and near-linear scaling. Proprioceptive states and exteroceptive observations (e.g., depth tokens) are encoded into compact tokens and fused by stacked SSD-Mamba2 layers. The selective state-space updates retain long-range dependencies with markedly lower latency and memory use than quadratic self-attention, enabling longer look-ahead, higher token resolution, and stable training under limited compute. Policies are trained end-to-end under curricula that randomize terrain and appearance and progressively increase scene complexity. A compact, state-centric reward balances task progress, energy efficiency, and safety. Across diverse motion-control scenarios, our approach consistently surpasses strong state-of-the-art baselines in return, safety (collisions and falls), and sample efficiency, while converging faster at the same compute budget. These results suggest that SSD-Mamba2 provides a practical fusion backbone for scalable, foresightful, and efficient end-to-end motion control.

Related papers

• DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation [69.8840101036735]
  Speculative decoding accelerates Large Language Model (LLM) inference.<n>Recent parallel drafters efficiently propose long token sequences in a single forward pass, but suffer from rapid acceptance decay due to a lack of inter-token dependencies.<n>We introduce DSpark, a speculative decoding framework that unifies high- throughput parallel generation with adaptive, load-aware verification.
  arXiv  Detail & Related papers  (2026-07-06T14:28:06Z)
• Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolation [15.129362213838974]
  We propose a trace-aware decoding framework with two components.<n>First, Temporal-Spatial Parallel Decoding (TSPD) uses a lightweight temporalspatial controller that consumes per-token trajectory features, including confidence, entropy, and momentum, together with token position, to decide when a token has converged and can be safely fixed.<n>Second, we introduce Confidence Extrapolation (CE), a training-free state-space module that forecasts future logit trends with uncertainty to support proactive decisions.
  arXiv  Detail & Related papers  (2026-05-29T02:29:28Z)
• TCP-SSM: Efficient Vision State Space Models with Token-Conditioned Poles [12.357939380869526]
  State Space Models (SSMs) have emerged as a compelling alternative to attention models for long-range vision tasks.<n>We propose Token-Conditioned Poles SSM, a structured selective SSM framework that improves efficiency while making recurrence dynamics explicit.<n>We show that TCP-SSM reduces SSM computation complexity up to 44% in Vision Mamba-style models while maintaining or surpassing baseline accuracy.
  arXiv  Detail & Related papers  (2026-05-12T05:49:46Z)
• DexWorldModel: Causal Latent World Modeling towards Automated Learning of Embodied Tasks [54.32016216994156]
  We introduce the Causal Latent World Model (CLWM), which employs DINOv3 features as generative targets to disentangle interaction semantics from visual noise.<n>To overcome memory scaling, CLWM features a Dual-State Test-Time Training (TTT) Memory that guarantees a strict $mathcalO(1)$ footprint for long-horizon tasks.<n>We present EmbodiChain, an online framework that establishes the Efficiency Law by injecting an infinite flow of physics-grounded trajectories during training.
  arXiv  Detail & Related papers  (2026-04-13T03:19:36Z)
• Bridging Large-Model Reasoning and Real-Time Control via Agentic Fast-Slow Planning [42.15812524999639]
  Agentic Fast-Slow Planning is a hierarchical framework that decouples perception, reasoning, planning, and control across natural timescales.<n>We show that Agentic Fast-Slow Planning improves robustness under perturbations, reducing lateral deviation by up to 45% and completion time by over 12% compared to pure MPC and an A*-guided MPC baseline.
  arXiv  Detail & Related papers  (2026-04-02T06:34:29Z)
• Beyond Short-Horizon: VQ-Memory for Robust Long-Horizon Manipulation in Non-Markovian Simulation Benchmarks [96.60530830276281]
  RuleSafe is a new articulated manipulation benchmark built upon a scalable LLM-aided simulation framework.<n>VQ-Memory is a compact and structured temporal representation that uses vector-quantized variational autoencoders.
  arXiv  Detail & Related papers  (2026-03-10T11:13:54Z)
• DriveMamba: Task-Centric Scalable State Space Model for Efficient End-to-End Autonomous Driving [47.573692944838115]
  DriveMamba is a Task-Centric Scalable paradigm for efficient E2E-AD.<n>It integrates sequential task relation modeling, implicit correspondence learning and long-term temporal fusion into a single-stage Unified Mamba decoder.<n>Extensive experiments conducted on nuScenes and Bench2Drive datasets demonstrate the superiority, generalizability and great efficiency of DriveMamba.
  arXiv  Detail & Related papers  (2026-02-09T11:48:29Z)
• Rethinking Multi-Condition DiTs: Eliminating Redundant Attention via Position-Alignment and Keyword-Scoping [61.459927600301654]
  Multi-condition control is bottlenecked by the conventional concatenate-and-attend'' strategy.<n>Our analysis reveals that much of this cross-modal interaction is spatially or semantically redundant.<n>We propose Position-aligned and Keyword-scoped Attention (PKA), a highly efficient framework designed to eliminate these redundancies.
  arXiv  Detail & Related papers  (2026-02-06T16:39:10Z)
• STEM: Scaling Transformers with Embedding Modules [59.26825251273227]
  We introduce STEM, a static, token-indexed approach that replaces the FFN up-projection with a layer-local embedding lookup.<n>This removes runtime routing, enables CPU offload with asynchronous prefetch, and decouples capacity from both per-token FLOPs and cross-device communication.<n>Overall, STEM is an effective way of scaling parametric memory while providing better interpretability, better training stability and improved efficiency.
  arXiv  Detail & Related papers  (2026-01-15T18:00:27Z)
• Training-free Context-adaptive Attention for Efficient Long Context Modeling [57.703159205740185]
  Training-free Context-adaptive Attention (TCA-Attention) is a training-free sparse attention mechanism that selectively attends to only the informative tokens for efficient long-context inference.<n>TCA-Attention achieves a 2.8$times$ speedup and reduces KV cache by 61% at 128K context length while maintaining performance comparable to full attention.
  arXiv  Detail & Related papers  (2025-12-10T01:54:57Z)
• Digital Twin-based Control Co-Design of Full Vehicle Active Suspensions via Deep Reinforcement Learning [8.45003507603975]
  Digital twins (DTs) and deep reinforcement learning (DRL) offer new opportunities for real-time, data-driven optimization across a vehicle's lifecycle.<n>This work presents a DT-based control co-design (CCD) framework for full-vehicle active suspensions.<n>We jointly optimize physical suspension components and control policies under varying driver behaviors and environmental uncertainties.
  arXiv  Detail & Related papers  (2025-12-03T15:41:35Z)
• Single Agent Robust Deep Reinforcement Learning for Bus Fleet Control [9.910562011343009]
  Bus bunching is a challenge for urban transit due to traffic and passenger demand.<n>We propose a novel single-agent reinforcement learning framework for bus holding control.<n>We show that our modified soft actor-critic achieves more stable and superior performance than benchmarks.
  arXiv  Detail & Related papers  (2025-08-28T13:47:40Z)
• LocoMamba: Vision-Driven Locomotion via End-to-End Deep Reinforcement Learning with Mamba [2.26464274357758]
  We introduce a vision-driven cross-modal DRL framework built on selective state-space models, specifically leveraging Mamba.<n>LocoMamba achieves near-linear-time sequence modeling, effectively captures long-range dependencies, and enables efficient training with longer sequences.<n>We evaluate our method in challenging simulated environments with static and moving obstacles as well as uneven terrain.
  arXiv  Detail & Related papers  (2025-08-16T00:13:24Z)
• SuperFlow++: Enhanced Spatiotemporal Consistency for Cross-Modal Data Pretraining [62.433137130087445]
  SuperFlow++ is a novel framework that integrates pretraining and downstream tasks using consecutive camera pairs.<n>We show that SuperFlow++ outperforms state-of-the-art methods across diverse tasks and driving conditions.<n>With strong generalizability and computational efficiency, SuperFlow++ establishes a new benchmark for data-efficient LiDAR-based perception in autonomous driving.
  arXiv  Detail & Related papers  (2025-03-25T17:59:57Z)
• DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Driving [62.62464518137153]
  DriveTransformer is a simplified E2E-AD framework for the ease of scaling up.<n>It is composed of three unified operations: task self-attention, sensor cross-attention, temporal cross-attention.<n>It achieves state-of-the-art performance in both simulated closed-loop benchmark Bench2Drive and real world open-loop benchmark nuScenes with high FPS.
  arXiv  Detail & Related papers  (2025-03-07T11:41:18Z)
• DiFSD: Ego-Centric Fully Sparse Paradigm with Uncertainty Denoising and Iterative Refinement for Efficient End-to-End Self-Driving [55.53171248839489]
  We propose an ego-centric fully sparse paradigm, named DiFSD, for end-to-end self-driving.<n>Specifically, DiFSD mainly consists of sparse perception, hierarchical interaction and iterative motion planner.<n>Experiments conducted on nuScenes and Bench2Drive datasets demonstrate the superior planning performance and great efficiency of DiFSD.
  arXiv  Detail & Related papers  (2024-09-15T15:55:24Z)
• An Efficient and Streaming Audio Visual Active Speaker Detection System [2.4515389321702132]
  We present two scenarios that address the key challenges posed by real-time constraints. First, we introduce a method to limit the number of future context frames utilized by the ASD model. Second, we propose a more stringent constraint that limits the total number of past frames the model can access during inference.
  arXiv  Detail & Related papers  (2024-09-13T17:45:53Z)
• VSSD: Vision Mamba with Non-Causal State Space Duality [26.96416515847115]
  State Space Models (SSMs) have gained prominence in vision tasks as they offer linear computational complexity. We introduce Visual State Space Duality (VSSD) model, which has a non-causal format of SSD. We conduct extensive experiments on various benchmarks including image classification, detection, and segmentation, where VSSD surpasses existing state-of-the-art SSM-based models.
  arXiv  Detail & Related papers  (2024-07-26T07:16:52Z)
• SHERL: Synthesizing High Accuracy and Efficient Memory for Resource-Limited Transfer Learning [63.93193829913252]
  We propose an innovative METL strategy called SHERL for resource-limited scenarios. In the early route, intermediate outputs are consolidated via an anti-redundancy operation. In the late route, utilizing minimal late pre-trained layers could alleviate the peak demand on memory overhead.
  arXiv  Detail & Related papers  (2024-07-10T10:22:35Z)
• Action-Quantized Offline Reinforcement Learning for Robotic Skill Learning [68.16998247593209]
  offline reinforcement learning (RL) paradigm provides recipe to convert static behavior datasets into policies that can perform better than the policy that collected the data. In this paper, we propose an adaptive scheme for action quantization. We show that several state-of-the-art offline RL methods such as IQL, CQL, and BRAC improve in performance on benchmarks when combined with our proposed discretization scheme.
  arXiv  Detail & Related papers  (2023-10-18T06:07:10Z)

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.