Searcharxiv⌕ Search

arXiv subjects

Weichao Zhuang

Publications and source records attributed to Weichao Zhuang.

13 recordsLinked to original sources

DynLimit: A Simulation Benchmark for Dynamics-Critical Planning and Control

Experiments near vehicle handling limits must keep three questions apart: what a maneuver requests, what the closed-loop system executes, and whether the task is completed. We present DynLimit, a simulation platform that organizes this process end to end. Scenarios are composed from road, friction, obstacle, and reference objects; planners and controllers attach through a shared observation and control interface; a vehicle adapter runs either a license-free four-wheel model or an external dynamics solver; and synchronized records feed task evaluation, replay, and response inspection. The platform carries a task-oriented evaluation method that pairs a prescribed reference demand with a sampled passing set and event-aligned diagnosis. Across eight task families, three friction environments, and three reference-assisted baseline chains we executed 450 runs: 360 nominal, 18 locally refined, and 72 with prescribed initial offsets. Method differences depend on the task family, equal aggregate pass counts arise from different passing cells, and local refinement narrows two sampled transitions without locating a continuous boundary. Two case comparisons show why sideslip or tire-utilization peaks cannot replace task constraints. The three views therefore stay distinct in every reported run: an outcome can be traced from the prescribed demand, through the executed response, to the task rule that decided it. Code is available at https://github.com/Yang2581/dynlimit

cs.RO↗

Lao Siji: Adapting openpilot to Closed-Loop End-to-End Driving in Neural Reconstruction

We present Lao Siji, an end-to-end driving policy that ranks third in the 2026 NVIDIA AlpaSim E2E Driving Challenge, ahead of most entries built on foundation models. Instead of training from scratch on a general-purpose vision backbone, we adapt openpilot, an open-source L2 production model. We freeze its vision encoder and fine-tune the policy layers with an ego-state adapter and a route memory. On closed-loop rollouts in neural-reconstructed scenes, we supervise the policy with a privileged map teacher for lateral self-correction and a body-extent loss for road and obstacle clearance, while distilling the base model's remaining outputs to retain car following. Ablations show that self-correction samples carry the gain, while the clearance loss substitutes for them where no teacher exists. Longitudinal safety instead rests on the base model's lead-vehicle estimate behind a rule-based speed layer. Longer training, imitating human logs, and human-log speed targets yield either no measurable gain or adverse effects. Without further training, Lao Siji also places among the leaders on HUGSIM and Odyssey. The final training takes only one GPU-hour and shifts the weights by 3.9% from the shipped model. Code will be released after review.

cs.RO↗

PlanWAM: Planning-Shaped Future Representations for End-to-End Autonomous Driving

World models in end-to-end autonomous driving predict future scene evolution to provide foresight for trajectory planning. Existing methods mainly study how to predict the future and how to use it, but less often ask which future representation is actually most useful for planning. To this end, we propose PlanWAM, a Planning-Shaped World Action Model. The key idea is to let the planning task shape the future-state representation, so that it retains the information most useful for planning. A latent world model then predicts this planning-shaped future latent representation from historical observations and uses it for planning, enabling foresighted planning. Specifically, we first use a Temporal Register Pyramid to compress multi-frame historical information in a recency-aware manner, learning a compact history representation oriented toward future reasoning and planning. We then introduce a privileged future posterior branch that observes ground-truth future frames, and shape its future latent representation with trajectory-planning objectives to obtain a planning-shaped future latent representation. Hindsight-to-Foresight Distillation trains a prior branch that depends only on history to predict this future latent representation. The predicted future latent representation serves as planning context and guides trajectory generation and selection. PlanWAM achieves 93.8 PDMS / 90.9 EPDMS on NAVSIM-v1/v2 navtest and reaches 38.7 HD-Score on closed-loop HUGSIM in a zero-shot setting, demonstrating leading planning performance across both open-loop and closed-loop evaluations. Extensive experiments further demonstrate that planning-shaped future representations provide an effective and deployable form of foresight for world-action models.

cs.RO↗

Risk-Aware Decision-Making for Autonomous Overtaking: A World Model-Based Mixture-of-Experts Framework

Autonomous highway overtaking demands foresighted decision-making to handle complex interactions, stochastic traffic evolution, and temporal risk accumulation. However, standard safe reinforcement learning approaches typically rely on implicit value-based risk estimations rather than explicit dynamics modeling, thereby struggling to accurately capture complex risk propagation over multi-step horizons. This limitation frequently results in behaviors that are locally safe but induce substantial latent risks in the long term. To address this, a World Model-based Risk-aware Mixture-of-Experts (WM-RMoE) framework is proposed. First, a learned latent dynamics model facilitates parallel multi-step rollouts, elevating safety assessment from the action level to the trajectory level via cumulative risk evaluation. Second, to enhance robustness under varying interaction intensities, a hierarchical gating mechanism dynamically coordinates experts across long-horizon, short-horizon, and rule-based safety modules. Furthermore, a Gaussian Mixture Model is integrated to preserve multimodal maneuvering branches, thereby mitigating the issue of behavioral mode averaging. Experimental results demonstrate that WM-RMoE significantly outperforms representative baselines in terms of safety compliance, decision stability, and generalization capability. Furthermore, benefiting from the risk-aware formulation, the proposed framework uniquely exhibits the ability to generate foresighted and semantically distinct overtaking maneuvers across diverse traffic densities.

cs.RO↗

LIDAR-AD: A Decoder-Free Latent-Interaction Dreamer with Action-Residual Chains for Autonomous Driving

Autonomous driving requires long-horizon closedloop decision making in dynamic traffic environments. Latent world models offer an effective framework for this problem by enabling imagination-based decision making in compact latent spaces. However, multi-source observations contain controlirrelevant redundancy, whereas reliable driving decisions rely on risk-relevant relations, future dynamics, and continuous action adjustments. This mismatch makes observation reconstruction and absolute action modeling suboptimal for learning decisionrelevant latent dynamics. We propose LIDAR-AD, a decoderfree Latent-Interaction Dreamer with Action-Residual Chains for autonomous driving. LIDAR-AD replaces observation reconstruction with redundancy-reduced latent alignment, encouraging compact representations of risk-relevant relations in multi-source driving inputs. It further models vehicle control as residual action updates and uses residual-action sequence contrastive learning to align multi-step residual-driven rollouts with future latent states. A deterministic analysis shows that the latent-tanh residual parameterization preserves interior action reachability while representing smooth long-horizon control as compact local updates. Together, these designs improve risk-aware state abstraction, continuous-control modeling, and long-horizon dynamics prediction. Extensive experiments across diverse simulated driving scenarios demonstrate that LIDAR-AD consistently outperforms world-model baselines, achieving the highest reward and the best success rate among learning-based methods. Evaluations on nuPlan-derived log-reconstructed scenarios further demonstrate the transferability of LIDAR-AD under real-world traffic layouts.

cs.LG↗

Large Language Models and Foundation Models in Smart Agriculture: Basics, Opportunities, and Challenges

The past decade has witnessed the rapid development and adoption of ML & DL methodologies in agricultural systems, showcased by great successes in agricultural applications. However, these conventional ML/DL models have certain limitations: they heavily rely on large, costly-to-acquire labeled datasets for training, require specialized expertise for development and maintenance, and are mostly tailored for specific tasks, thus lacking generalizability. Recently, large pre-trained models, also known as FMs, have demonstrated remarkable successes in language, vision, and decision-making tasks across various domains. These models are trained on a large amount of data from multiple domains and modalities. Once trained, they can accomplish versatile tasks with just minor fine-tuning and minimal task-specific labeled data. Despite their proven effectiveness and huge potential, there has been little exploration of applying FMs to agriculture AI. Thus, this study aims to explore the potential of FMs in the field of smart agriculture. In particular, conceptual tools and technical background are presented to help the understanding of the problem space and uncover new research directions. To this end, recent FMs in the general CS domain are reviewed, and the models are categorized into four categories: language FMs, vision FMs, multimodal FMs, and reinforcement learning FMs. Then, the steps of developing agriculture FMs (AFMs) are outlined and potential applications in smart agriculture are discussed. Moreover, challenges and risks associated with developing AFMs are discussed, including model training, validation, and deployment. In summary, the advancement of AI in agriculture is explored by introducing AFMs as a promising paradigm that can significantly mitigate the reliance on extensive labeled datasets and enhance the efficiency, effectiveness, and generalization of agricultural AI systems.

cs.LG↗

SoybeanNet: Transformer-Based Convolutional Neural Network for Soybean Pod Counting from Unmanned Aerial Vehicle (UAV) Images

Soybeans are a critical source of food, protein and oil, and thus have received extensive research aimed at enhancing their yield, refining cultivation practices, and advancing soybean breeding techniques. Within this context, soybean pod counting plays an essential role in understanding and optimizing production. Despite recent advancements, the development of a robust pod-counting algorithm capable of performing effectively in real-field conditions remains a significant challenge This paper presents a pioneering work of accurate soybean pod counting utilizing unmanned aerial vehicle (UAV) images captured from actual soybean fields in Michigan, USA. Specifically, this paper presents SoybeanNet, a novel point-based counting network that harnesses powerful transformer backbones for simultaneous soybean pod counting and localization with high accuracy. In addition, a new dataset of UAV-acquired images for soybean pod counting was created and open-sourced, consisting of 113 drone images with more than 260k manually annotated soybean pods captured under natural lighting conditions. Through comprehensive evaluations, SoybeanNet demonstrated superior performance over five state-of-the-art approaches when tested on the collected images. Remarkably, SoybeanNet achieved a counting accuracy of $84.51\%$ when tested on the testing dataset, attesting to its efficacy in real-world scenarios. The publication also provides both the source code (\url{https://github.com/JiajiaLi04/Soybean-Pod-Counting-from-UAV-Images}) and the labeled soybean dataset (\url{https://www.kaggle.com/datasets/jiajiali/uav-based-soybean-pod-images}), offering a valuable resource for future research endeavors in soybean pod counting and related fields.

cs.CV↗

Challenges and Opportunities for Second-life Batteries: A Review of Key Technologies and Economy

Due to the increasing volume of Electric Vehicles in automotive markets and the limited lifetime of onboard lithium-ion batteries (LIBs), the large-scale retirement of LIBs is imminent. The battery packs retired from Electric Vehicles still own 70%-80% of the initial capacity, thus having the potential to be utilized in scenarios with lower energy and power requirements to maximize the value of LIBs. However, spent batteries are commonly less reliable than fresh batteries due to their degraded performance, thereby necessitating a comprehensive assessment from safety and economic perspectives before further utilization. To this end, this paper reviews the key technological and economic aspects of second-life batteries (SLBs). Firstly, we introduce various degradation models for first-life batteries and identify an opportunity to combine physics-based theories with data-driven methods to establish explainable models with physical laws that can be generalized. However, degradation models specifically tailored to SLBs are currently absent. Therefore, we analyze the applicability of existing battery degradation models developed for first-life batteries in SLB applications. Secondly, we investigate fast screening and regrouping techniques and discuss the regrouping standards for the first time to guide the classification procedure and enhance the performance and safety of SLBs. Thirdly, we scrutinize the economic analysis of SLBs and summarize the potentially profitable applications. Finally, we comprehensively examine and compare power electronics technologies that can substantially improve the performance of SLBs, including high-efficiency energy transformation technologies, active equalization technologies, and technologies to improve reliability and safety.

eess.SY↗

Overtaking-enabled Eco-approach Control at Signalized Intersections for Connected and Automated Vehicles

Preceding vehicles typically dominate the movement of following vehicles in traffic systems, thereby significantly influencing the efficacy of eco-driving control that concentrates on vehicle speed optimization. To potentially mitigate the negative effect of preceding vehicles on eco-driving control at the signalized intersection, this paper proposes an overtakingenabled eco-approach control (OEAC) strategy. It combines driving lane planning and speed optimization for connected and automated vehicles to relax the first-in-first-out queuing policy at the signalized intersection, minimizing the target vehicle's energy consumption and travel delay. The OEAC adopts a receding horizon two-stage control framework to derive optimal driving trajectories for adapting to dynamic traffic conditions. In the first stage, the driving lane optimization problem is formulated as a Markov decision process and solved using dynamic programming, which takes into account the uncertain disturbance from preceding vehicles. In the second stage, the vehicle's speed trajectory with the minimal driving cost is optimized rapidly using Pontryagin's minimum principle to obtain the closed-form analytical optimal solution. Extensive simulations are conducted to evaluate the effectiveness of the OEAC. The results show that the OEAC is excellent in driving cost reduction over constant speed and regular eco-approach and departure strategies in various traffic scenarios, with an average improvement of 20.91% and 5.62%, respectively.

eess.SY↗

Adaptive Leading Cruise Control in Mixed Traffic Considering Human Behavioral Diversity

This paper presents an adaptive leading cruise control strategy for the connected and automated vehicle (CAV) and first considers its impact on the following human-driven vehicle (HDV) with diverse driving characteristics in the unified optimization framework for improved holistic energy efficiency. The car-following behaviors of HDV are statistically calibrated using the Next Generation Simulation dataset. In a typical single-lane car-following scenario where CAVs and HDVs share the road, the longitudinal speed control of CAVs can substantially reduce the energy consumption of the following HDV by avoiding unnecessary acceleration and braking. Moreover, apart from the objectives including car-following safety and traffic efficiency, the energy efficiencies of both CAV and HDV are incorporated into the reward function of reinforcement learning. The specific driving pattern of the following HDV is learned in real-time from historical speed information to predict its acceleration and power consumption in the optimization horizon. A comprehensive simulation is conducted to statistically verify the positive impacts of CAV on the holistic energy efficiency of the mixed traffic flow with uncertain and diverse human driving behaviors. Simulation results indicate that the holistic energy efficiency is improved by 4.38% on average.

eess.SY↗

Cloud-Assisted Collaborative Road Information Discovery with Gaussian Process: Application to Road Profile Estimation

There is an increasing popularity in exploiting modern vehicles as mobile sensors to obtain important road information such as potholes, black ice and road profile. Availability of such information has been identified as a key enabler for next-generation vehicles with enhanced safety, efficiency, and comfort. However, existing road information discovery approaches have been predominately performed in a single-vehicle setting, which is inevitably susceptible to vehicle model uncertainty and measurement errors. To overcome these limitations, this paper presents a novel cloud-assisted collaborative estimation framework that can utilize multiple heterogeneous vehicles to iteratively enhance estimation performance. Specifically, each vehicle combines its onboard measurements with a cloud-based Gaussian process (GP), crowdsourced from prior participating vehicles as "pseudo-measurements", into a local estimator to refine the estimation. The resultant local onboard estimation is then sent back to the cloud to update the GP, where we utilize a noisy input GP (NIGP) method to explicitly handle uncertain GPS measurements. We employ the proposed framework to the application of collaborative road profile estimation. Promising results on extensive simulations and hardware-in-the-loop experiments show that the proposed collaborative estimation can significantly enhance estimation and iteratively improve the performance from vehicle to vehicle, despite vehicle heterogeneity, model uncertainty, and measurement noises.

eess.SY↗

Decentralized On-Ramp Merging Control of Connected and Automated Vehicles in the Mixed Traffic Using Control Barrier Functions

The cooperative control of the connected and automated vehicle (CAV) is recognized as an effective approach to alleviate traffic congestion and improve traffic safety, especially for on-ramp bottlenecks. However, in the mixed traffic, the uncertainty of human-driven vehicles (HDVs) makes the on-ramp merging control for CAVs more challenging. This paper proposes a decentralized optimal control method to address the merging control problem of CAVs at highway on-ramps in the mixed traffic. We first formulate the optimal merging control problem, which includes the constraints of safety and vehicle dynamics, with the objectives of minimizing travel time and energy consumption. Then, a control framework, combining control barrier functions (CBFs) and control Lyapunov functions (CLFs) is proposed. CBFs render the system subject to safety-critical constraints, while CLFs stabilize the system to the objectives. In addition, to enable effective control of CAVs in the mixed traffic, a recursive merging control framework is proposed, where HDVs are regarded as disturbances, and CAVs collect surrounding vehicles' states repeatedly and update their trajectories recursively to satisfy strict merging requirements. Finally, the merging problem is reformulated as a quadratic programming problem, which allows for real-time application. Simulation results show that the proposed on-ramp merging control method is robust in resisting disturbance from the HDV with traffic efficiency and energy economy improvement.

eess.SY↗

Optimal design of three-planetary-gear power-split hybrid powertrains

Many of today's power-split hybrid electric vehicles (HEVs) utilize planetary gears (PGs) to connect the powertrain elements together. Recent power-split HEVs tend to use two PGs and some of them have multiple modes to achieve better fuel economy and driving performance. Looking to the future, hybrid powertrain technologies must be enhanced to design hybrid light trucks. For light trucks, the need for multi-mode and more PGs is stronger, to achieve the required performance. To systematically explore all the possible designs of multi-mode HEVs with three PGs, an efficient searching and optimization methodology is proposed. All possible clutch topology and modes for one existing configuration that uses three PGs were exhaustively searched. The launching performance is first used to screen out designs that fail to satisfy the required launching performance. A near-optimal and computationally efficient energy management strategy was then employed to identify designs that achieve good fuel economy. The proposed design process successfully identify 8 designs that achieve better launching performance and better fuel economy, while using fewer number of clutches than the benchmark and a patented design.

eess.SY↗