We propose an energy-based control method for a multilocomotion robot to improve the stability of continuous brachiation. The target continuous brachiation is an effective locomotion from one bar to another exchanging a kinetic energy with potential energy like a pendulum. Our control strategy for the continuous brachiation is to control the swing-back action so that the robot pumps up the feasible energy to grasp a target bar, and then the locomotion action is designed based on the symmetric motion of a pendulum in order to conserve the total energy: summation of the kinetic and potential energy. The proposed controller is implemented on a new type of mobile multilocomotion robot named “Gorilla Robot III” whose dimensions are mimicking those of a gorilla. Continuous brachiation on various uniform ladders is experimentally achieved with this robot. The experimental results show the validity of our control algorithm. © 2006 Wiley Periodicals, Inc. Int J Int Syst 21: 1025–1043, 2006.

In this paper, we present a control method to realize continuous brachiation. The target brachiation is basically divided into two actions: a swing action and a locomotion action. In order to realize the continuous brachiation effectively and stably, it is necessary to control these actions according to the total energy of the system. Because brachiation has been modeled as a pendulum-like motion and the amplitude of the oscillation determines whether the robot can grasp the target bar or not. In consideration of this point, an energy based control method is proposed and introduced into the swing action controller. The purpose of this method is to control the angular velocity of the swing action in order to satisfy the desired energy evaluated by the distance of bars. Experimental results show that the robot can successfully achieve smooth, continuous brachiation.

The goal of our research is to achieve a dynamic and skillful motion by a real robot. We have previously developed a brachiation controller that can achieve two kinds of brachiation: overhand and side-hand motions. In this paper, we redesign behavior controllers in order to achieve a stable continuous brachiation without a break. We make the behavior controllers that can be controlled by the angle around a catching bar because the angle is the most dominant parameters of the brachiation motions. In addition, we introduce the action of bending elbow joints in order to realize a pendulum motion and to reduce the impact of the moment of catching a bar. Finally, we show the experimental results that the robot could successfully achieve continuous brachiation.

In this paper, we firstly introduce new concept of multi-locomotion robot inspired by an animal and a developed robot "Gorilla Robot II", which can select a better locomotion from biped locomotion, quadruped locomotion and brachiation, according to an environment. We consider "Brachiation" which is one of the most dynamic motions in animal motions. Then we propose an enhanced control method for the robot in order to achieve a brachiation motion, adjusting the timing of local behaviors. We show that the developed robot successfully perform two kinds of brachiation motions.