High free volume, film-forming copolymers were prepared in which a proportion of the spirounits of PIM-1 were replaced with units derived from 9,10-dimethyl-9,10-dihydro-9,10-ethanoanthracene-2,3,6,7-tetrol (CO1). A full investigation of free volume, utilizing N 2 sorption, positron annihilation lifetime spectroscopy (PALS), Xe sorption and 129 Xe NMR spectroscopy, was undertaken for copolymer PIM1-CO1-40 (spiro-units:CO1 = 60:40) and a comparison is made with PIM-1. All techniques indicate that the copolymer, like PIM-1, possesses free volume holes or pores on the nanometre length scale (i.e., microporosity as defined by IUPAC). For the batch of PIM-1 studied here, the sample as received showed anomalous N 2 sorption, Xe sorption and 129 Xe NMR behavior that could be interpreted in terms of reduced porosity in the size range 0.6-0.7 nm, as compared to the copolymer. The anomalous behavior was eliminated on conditioning or relaxation of the polymer, e.g., by Xe sorption at 100 °C and 3 bar. PALS for both PIM1-CO1-40 and PIM-1 indicates a maximum in the average free volume hole size, and in the width of the distribution of hole sizes, on increasing temperature. This maximum appears to be a feature of high free volume polymers and may be related to the onset of localized oscillations of backbone moieties.

We present a Learning from Demonstration method for teaching robots to perform search strategies imitated from humans in scenarios where alignment tasks fail due to position uncertainty. The method utilizes human demonstrations to learn both a state invariant dynamics model and an exploration distribution that captures the search area covered by the demonstrator. We present two alternative algorithms for computing a search trajectory from the exploration distribution, one based on sampling and another based on deterministic ergodic control. We augment the search trajectory with forces learnt through the dynamics model to enable searching both in force and position domains. An impedance controller with superposed forces is used for reproducing the learnt strategy. We experimentally evaluate the method on a KUKA LWR4+ performing a 2D peg-in-hole and a 3D electricity socket task. Results show that the proposed method can, with only few human demonstrations, learn to complete the search task.

No abstract