arXiv subjects

Yipeng Yang

Publications and source records attributed to Yipeng Yang.

12 recordsLinked to original sources

arXiv2026

Preserving Full 6-DOF Actuation Under Abrupt Total Rotor Failures: Passive Fault-Tolerant Flight Control Using a Biaxial-Tilt Hexacopter

Conventional multirotors suffer from a rapid collapse of attainable wrench space (AWS) under abrupt total rotor failures, rendering full 6-DOF recovery physically impossible. This paper addresses passive fault-tolerant flight of a biaxial-tilt overactuated hexacopter (BTO) under abrupt total rotor failures that are a priori unknown to the controller. The control design and analysis focus on representative abrupt rotor-failure cases for which the post-failure system remains fully actuated, while no explicit fault detection, isolation, or fault-mode switching is assumed. First, we extend the inscribed-sphere metric of the AWS by incorporating the transient-wrench-jump term, enabling quantitative feasibility assessment under up to three simultaneous rotor failures and benchmarking against uniaxial-tilt and coplanar hexacopters. Second, we develop two computationally efficient passive schemes without relying on fault detection or online optimization. One scheme operates at the controller layer by combining a high-order fully actuated (HOFA) controller with a linear extended state observer (LESO) for lumped-disturbance rejection. The other scheme operates at the allocator layer by using model-reference adaptive control allocation with momentum-based wrench estimation to compensate for control-allocation biases. Simulations and flight experiments validate stable hovering and 6-DOF trajectory tracking under single and multiple rotor failures. Further systematic comparisons confirm that the BTO provides larger recovery margins than uniaxial-tilt and coplanar designs. Additional onboard-sensor-only experiments, including indoor tracking under wind disturbance, outdoor tracking under extreme conditions, narrow-frame traversal, and contact-based aerial writing, further validate the robustness of the proposed framework in complex operational environments.

cs.RO↗

arXiv2026

Attitude-Aided Linear Calibration of Triaxial Accelerometers

Triaxial MEMS accelerometers are widely used for inertial sensing, navigation, and sensor fusion, but existing calibration methods often rely on costly reference setups or nonlinear iterative optimization, limiting their efficiency and applicability to low-cost or self-calibrating systems. We present attitude-aided linear accelerometer calibration (ALAC), a method that operates on any platform providing orientation information, such as turntables, robotic arms, or inertial measurement units. ALAC constructs a combined error matrix (CEM) to represent sensor errors in a unified calibration model and enables linear least-squares estimation. The bias and gravity vector are jointly estimated, implicitly accounting for platform misalignment, and matrix decomposition of the CEM recovers scale, non-orthogonality, and alignment rotation parameters. Under static gravity, calibration is formulated as a constrained homogeneous least-squares (CHLS) problem and solved in closed form using standard linear algebra. Only five arbitrarily oriented measurements are required, and a recursive extension supports online or in-field calibration. Experiments on a stationary robot-mounted accelerometer and a quasi-static public IMU trajectory show that ALAC, in both offline and online modes, outperforms reference-based and online baselines in accuracy and robustness to sensor noise. On the same dataset, it matches iterative self-calibration under filtered conditions and surpasses all evaluated baselines on raw measurements. These results demonstrate a robust and practical calibration scheme for MEMS-based inertial platforms, especially low-cost IMUs and online calibration scenarios.

cs.RO↗

arXiv2026

Truncated Rectified Flow Policy for Reinforcement Learning with One-Step Sampling

Maximum entropy reinforcement learning (MaxEnt RL) has become a standard framework for sequential decision making, yet its standard Gaussian policy parameterization is inherently unimodal, limiting its ability to model complex multimodal action distributions. This limitation has motivated increasing interest in generative policies based on diffusion and flow matching as more expressive alternatives. However, incorporating such policies into MaxEnt RL is challenging for two main reasons: the likelihood and entropy of continuous-time generative policies are generally intractable, and multi-step sampling introduces both long-horizon backpropagation instability and substantial inference latency. To address these challenges, we propose Truncated Rectified Flow Policy (TRFP), a framework built on a hybrid deterministic-stochastic architecture. This design makes entropy-regularized optimization tractable while supporting stable training and effective one-step sampling through gradient truncation and flow straightening. Empirical results on a toy multigoal environment and 10 MuJoCo benchmarks show that TRFP captures multimodal behavior effectively, outperforms strong baselines on most benchmarks under standard sampling, and remains highly competitive under one-step sampling.

cs.LG↗

arXiv2024

ATDM:An Anthropomorphic Aerial Tendon-driven Manipulator with Low-Inertia and High-Stiffness

Aerial Manipulator Systems (AMS) have garnered significant interest for their utility in aerial operations. Nonetheless, challenges related to the manipulator's limited stiffness and the coupling disturbance with manipulator movement persist. This paper introduces the Aerial Tendon-Driven Manipulator (ATDM), an innovative AMS that integrates a hexrotor Unmanned Aerial Vehicle (UAV) with a 4-degree-of-freedom (4-DOF) anthropomorphic tendon-driven manipulator. The design of the manipulator is anatomically inspired, emulating the human arm anatomy from the shoulder joint downward. To enhance the structural integrity and performance, finite element topology optimization and lattice optimization are employed on the links to replicate the radially graded structure characteristic of bone, this approach effectively reduces weight and inertia while simultaneously maximizing stiffness. A novel tensioning mechanism with adjustable tension is introduced to address cable relaxation, and a Tension-amplification tendon mechanism is implemented to increase the manipulator's overall stiffness and output. The paper presents a kinematic model based on virtual coupled joints, a comprehensive workspace analysis, and detailed calculations of output torques and stiffness for individual arm joints. The prototype arm has a total weight of 2.7 kg, with the end effector contributing only 0.818 kg. By positioning all actuators at the base, coupling disturbance are minimized. The paper includes a detailed mechanical design and validates the system's performance through semi-physical multi-body dynamics simulations, confirming the efficacy of the proposed design.

cs.RO↗

arXiv2023

A Microgrid Trading Framework Based on PoC Consensus

In the field of energy Internet, blockchain-based distributed energy trading mode is a promising way to replace the traditional centralized trading mode. However, the current power blockchain platform based on public chain has problems such as low consensus efficiency and waste of computing resources. The energy trading platform based on the consortium chain has problems such as inability to attract users to join and collusion between alliances. We propose a microgrid trading framework based on proof of contribution (PoC). According to the contribution value, we randomly select nodes based on weights through verifiable random functions (VRF) in each round of consensus to form the next round of consensus committee. The proposed electricity trading framework not only improves the consensus efficiency of the blockchain, but also is suitable as an incentive mechanism to attract users to participate in the power blockchain. Experiments show that our framework is effective and the contribution value we designed is reasonable. Through our framework, we can motivate users to participate and make energy transactions more fair.

cs.DC↗

arXiv2014

Finite Horizon Time Inhomogeneous Singular Control Problem of One-dimensional Diffusion via Dynkin Game

The Hamilton-Jacobi-Bellman equation (HJB) associated with the time inhomogeneous singular control problem is a parabolic partial differential equation, and the existence of a classical solution is usually difficult to prove. In this paper, a class of finite horizon stochastic singular control problems of one dimensional diffusion is solved via a time inhomogeneous zero-sum game (Dynkin game). The regularity of the value function of the Dynkin game is investigated, and its integrated form coincides with the value function of the singular control problem. We provide conditions under which a classical solution to the associated HJB equation exists, thus the usual viscosity solution approach is avoided. We also show that the optimal control policy is to reflect the diffusion between two time inhomogeneous boundaries. For a more general terminal payoff function, we showed that the optimal control involves a possible impulse at maturity.

math.OC↗

arXiv2014

Multi-dimensional Stochastic Singular Control Via Dynkin Game and Dirichlet Form

The traditional difficulty about stochastic singular control is to characterize the regularities of the value function and the optimal control policy. In this paper, a multi-dimensional singular control problem is considered. We found the optimal value function and the optimal control policy of this problem via Dynkin game, whose solution is given by the saddle point of the cost function. The existence and uniqueness of the solution to this Dynkin game are proved through an associated variational inequality problem involving Dirichlet form. As a consequence, the properties of the value function of this Dynkin game implies the smoothness of the value function of the stochastic singular control problem. In this way, we are able to show the existence of a classical solution to this multi-dimensional singular control problem, which was traditionally solved in the sense of viscosity solutions, and this enables the application of the verification theorem to prove optimality.

math.OC↗

arXiv2014

Prospect Agents and the Feedback Effect on Price Fluctuations

A microeconomic approach is proposed to derive the fluctuations of risky asset price, where the market participants are modeled as prospect trading agents. As asset price is generated by the temporary equilibrium between demand and supply, the agents' trading behaviors can affect the price process in turn, which is called the feedback effect. The prospect agents make actions based on their reactions to gains and losses, and as a consequence of the feedback effect, a relationship between the agents' trading behavior and the price fluctuations is constructed, which explains the implied volatility skew and smile observed in actual market.

q-fin.PR↗

arXiv2013

A Correction to "Dynkin Games Via Dirichlet Forms and Singular Control of One-Dimensional Diffusion"

In the paper "Dynkin Games Via Dirichlet Forms and Singular Control of One-Dimensional Diffusion", the authors tried to show the existences of a smooth value function and an optimal policy to a one-dimensional stochastic singular control problem, where the underlying process is a generalized diffusion process given by dX_t=μ(X_t)dt+σ(X_t)dw_t, in which w_t is a Wiener process. It is found that either a condition on μ(x) and σ(x) should be added, or a different diffusion process should be considered and as a result, the main theorem of this paper should be amended.

math.OC↗

arXiv2013

Refined Solutions of Time Inhomogeneous Optimal Stopping Games via Dirichlet Form

The properties of value functions of time inhomogeneous optimal stopping problem and zero-sum game (Dynkin game) are studied through time dependent Dirichlet form. Under the absolute continuity condition on the transition function of the underlying diffusion process and some other assumptions, the refined solutions without exceptional starting points are proved to exist, and the value functions of the optimal stopping and zero-sum game, which are finely and cofinely continuous, are characterized as the solutions of some variational inequalities, respectively.

math.OC↗

arXiv2013

A Monte Carlo Simulation on Clustering Dynamics of Social Amoebae

A discrete model for computer simulations of the clustering dynamics of Social Amoebae is presented. This model incorporates the wavelike propagation of extracellular signaling cAMP, the sporadic firing of cells at early stage of aggregation, the signal relaying as a response to stimulus, the inertia and purposeful random walk of the cell movement. A Monte Carlo simulation is run which shows the existence of potential equilibriums of mean and variance of aggregation time. The simulation result of this model could well reproduce many phenomena observed by actual experiments.

q-bio.CB↗

arXiv2013

On the paradox of pesticides

The paradox of pesticides was observed experimentally, which says that pesticides may dramatically increase the population of a pest when the pest has a natural predator. Here we use a mathematical model to study the paradox. We find that the timing for the application of pesticides is crucial for the resurgence or non-resurgence of the pests. In particular, regularly applying pesticides is not a good idea as also observed in experiments. In fact, the best time to apply pesticides is when the pest population is reasonably high.

q-bio.PE↗