arXiv subjects

Koju Nishimoto

Publications and source records attributed to Koju Nishimoto.

3 recordsLinked to original sources

arXiv2026

A Controllability Gramain Shaping with LMI Constraints under Bures--Wasserstein Distance

This paper proposes a controller design method for shaping the controllability Gramian into a desired form to design the effect from exogenous inputs to the system state. Using the Bures--Wasserstein distance, we formulate the shaping problem as the minimization of the distance between the system Gramian and a desired Gramian, and the objective function is shown to be strictly convex on the set of symmetric positive definite matrices. In addition, by deriving a semidefinite programming formulation via a linear matrix inequality (LMI), computational efficiency is improved and additional LMI constraints can be incorporated. When the exogenous input is modeled as Gaussian white noise, the proposed framework is closely related to $H_2$ control, which can be interpreted as a special case of optimal transport. Numerical examples demonstrate anisotropic controllability design for a guidance robot and verify the ability to impose additional directional constraints through LMIs. The numerical examples also confirm that the proposed method approaches $H_2$ control as the desired Gramian tends to zero.

math.OC↗

arXiv2023

Collision Avoidance for Ellipsoidal Rigid Bodies with Control Barrier Functions Designed from Rotating Supporting Hyperplanes

This paper proposes a collision avoidance method for ellipsoidal rigid bodies, which utilizes a control barrier function (CBF) designed from a supporting hyperplane. We formulate the problem in the Special Euclidean Group SE(2) and SE(3), where the dynamics are described as rigid body motion (RBM). Then, we consider the condition for separating two ellipsoidal rigid bodies by employing a signed distance from a supporting hyperplane of a rigid body to the other rigid body. Although the positive value of this signed distance implies that two rigid bodies are collision-free, a naively prepared supporting hyperplane yields a smaller value than the actual distance. To avoid such a conservative evaluation, the supporting hyperplane is rotated so that the signed distance from the supporting hyperplane to the other rigid body is maximized. We prove that the maximum value of this optimization problem is equal to the actual distance between two ellipsoidal rigid bodies, hence eliminating excessive conservativeness. We leverage this signed distance as a CBF to prevent collision while the supporting hyperplane is rotated via a gradient-based input. The designed CBF is integrated into a quadratic programming (QP) problem, where each rigid body calculates its collision-free input in a distributed manner, given communication among rigid bodies. The proposed method is demonstrated with simulations. Finally, we exemplify our method can be extended to a vehicle having nonholonomic dynamics.

eess.SY↗

arXiv2022

Collision Avoidance for Elliptical Agents with Control Barrier Function Utilizing Supporting Lines

This paper presents a collision avoidance method for elliptical agents traveling in a two-dimensional space. We first formulate a separation condition for two elliptical agents utilizing a signed distance from a supporting line of an agent to the other agent, which renders a positive value if two ellipses are separated by the line. Because this signed distance could yield a shorter length than the actual distance between two ellipses, the supporting line is rotated so that the signed distance from the line to the other ellipse is maximized. We prove that this maximization problem renders the signed distance equivalent to the actual distance between two ellipses, hence not causing the conservative evasive motion. Then, we propose the collision avoidance method utilizing novel control barrier functions incorporating a gradient-based update law of a supporting line. The validity of the proposed methods is evaluated in the simulations.

eess.SY↗