In this study, we propose a triaxial force measurement sensor probe with piezoresistors fabricated via sidewall doping using rapid thermal diffusion. The device was developed as a tool for measuring micronewton-level forces as vector quantities. The device consists of a 15 µm thick cantilever, two sensing beams and four wiring beams. The length and width of the cantilever are 1240 µm and 140 µm, respectively, with a beam span of 1200 µm and a width of 10–15 µm. The piezoresistors are formed at the root of the cantilever and the sidewalls of the two sensing beams. The sensor spring constants for each axis were measured at kx = 1.5 N m−1, ky = 3.5 N m−1 and kz = 0.64 N m−1. We confirmed that our device was capable of measuring triaxial forces with a minimum detectable force at the submicronewton level.

This paper presents direct measurements of the aerodynamic forces on the wing of a free-flying, insect-like ornithopter that was modeled on a hawk moth (Manduca sexta). A micro differential pressure sensor was fabricated with micro electro mechanical systems (MEMS) technology and attached to the wing of the ornithopter. The sensor chip was less than 0.1% of the wing area. The mass of the sensor chip was 2.0 mg, which was less than 1% of the wing mass. Thus, the sensor was both small and light in comparison with the wing, resulting in a measurement system that had a minimal impact on the aerodynamics of the wing. With this sensor, the 'pressure coefficient' of the ornithopter wing was measured during both steady airflow and actual free flight. The maximum pressure coefficient observed for steady airflow conditions was 1.4 at an angle of attack of 30 degrees . In flapping flight, the coefficient was around 2.0 for angles of attack that ranged from 25 degrees to 40 degrees . Therefore, a larger aerodynamic force was generated during the downstroke in free flight compared to steady airflow conditions.

No abstract

Differential Dynamic Programming (DDP) has become a well established method for unconstrained trajectory optimization. Despite its several applications in robotics and controls however, a widely successful constrained version of the algorithm has yet to be developed. This paper builds upon penalty methods and active-set approaches, towards designing a Dynamic Programming-based methodology for constrained optimal control. Regarding the former, our derivation employs a constrained version of Bellman's principle of optimality, by introducing a set of auxiliary slack variables in the backward pass. In parallel, we show how Augmented Lagrangian methods can be naturally incorporated within DDP, by utilizing a particular set of penalty-Lagrangian functions that preserve second-order differentiability. We demonstrate experimentally that our extensions (individually and combinations thereof) enhance significantly the convergence properties of the algorithm, and outperform previous approaches on a large number of simulated scenarios.

No abstract