Abstract
This paper proposes a novel dual-arm motion discrimination method combining of the posterior probabilities estimated independently for left and right arm motions. In the proposed method, first, only the posterior probability of each single-arm motion is estimated through learning from measured biological signals using recurrent probabilistic neural networks. The posterior probabilities output from the recurrent probabilistic neural networks are then combined based on motion dependency between arms, making it possible to calculate a joint posterior probability of dual-arm motions. With this method, all the dual-arm motions consisting of each single-arm motion can be discriminated through leaning of single-arm motions only. In the experiments, the proposed method was applied to discrimination of 15 dual-arm motions which consist of three right-arm motions, three left-left arm motions and nine combined dual-arm motions. The results showed that the proposed method could achieve high discrimination performance though leaning of three motions for each arm only (average discrimination rates: 97.49±2.37%). In addition, the possibility of applying the proposed method for a human interface was confirmed through operation experiments for the glovebox system.
