The objective of the development of Treebot is to assist or replace human being in performing forestry tasks on trees. Treebot is composed of a pair of tree grippers that permits Treebot to attach on a wide variety of trees with a wide range of gripping curvature, and a novel continuum maneuvering structure that permits Treebot has high maneuverability and adaptability. In the robot actuation, merely five actuators are necessary. Although Treebot weighs 600grams only, it has payload capability of 1.75kg which is nearly three times of its own weight.
A certain level of autonomous climbing ability of Treebot is helpful to reduce the complexity of manipulation by users. An autonomous climbing strategy for Treebot is thus proposed. To determine the motions to climb up autonomously in an unknown environment, a robot must be equipped with sensors that can explore the environment. Vision-based sensors provide rich information about the environment. However, they require a great deal of computational power. Moreover, light conditions vary in outdoor environments, which may affect the accuracy of visual information. There are many living creatures that do not rely on visual information, but can navigate well in their natural environment. Inchworms, for example, navigate on trees by using only their sense of touch. Although the information obtained by tactile sensors is not rich, it is reliable. Furthermore, the processing of tactile information is much simpler than that of visual information.
Inspired by arboreal animals, an algorithm is developed to allow Treebot to climb irregularly shaped trees autonomously by using tactile sensors and a tilting sensor only. The development of the algorithm can also reveal how tactile sensors can best be employed in autonomous tree climbing.