We present work on a robot system capable of rapidly learning to avoid obstacles in previously unseen environments. We exploit deep learning’s generalized features to reason about when to turn left, turn right, or drive straight within an indoor environment. We present preliminary results of a deep convolutional neural network trained without any pre-training that can successfully avoid obstacles.

A new real-time obstacle avoidance approach for mobile robots has been developed and implemented. This approach permits the detection of unknown obstacles simultaneously with the steering of the mobile robot to avoid collisions and advancing toward the target. The novelty of this approach, entitled the Virtual Force Field, lies in the integration of two known concepts: Certainty Grids for obsta...

In this paper, we provide a dynamic graph theoretical framework for flocking in presence of multiple obstacles. In particular, we give formal definitions of nets and flocks as spatially induced graphs. We provide models of nets and flocks and discuss the realization/embedding issues related to structural nets and flocks. This allows task representation and execution for a network of agents call...

This paper describes a laser ranging sensor that is suitable for applications like small unmanned aerial vehicles. The hardware consists of a diode emitter array and line-scan charge coupled devices. A structured-light technique measures ranges up to 30 meters for 64 field angles in a 90 degree field of view. Operation is eye safe, and the laser wavelength is not visible to night vision goggles...

The Stanford AI Lab cart is a remotely controlled TV equipped mobile robot. A computer program has driven the cart through cluttered spaces, gaining its knowledge of the world entirely from images broadcast by the onboard TV system. The cart uses several kinds of stereo to locate objects around it in 3D and to deduce its own motion. It plans an obstacle avoiding path to a desired destination on...

Compared with the space fixed feature of traditional wireless sensor network (WSN), mobile WSN has better robustness and adaptability in unknown environment, so that it is always applied in the research of target tracking. In order to reach the target, the nodes group should find a self-adaptive method to avoid the obstacles together in their moving directions. Previous methods, which were base...

In this paper, we present an approach to obstacle avoidance for a group of unmanned vehicles moving in formation. The goal of the group is to move through a partially unknown environment with obstacles and reach a destination while maintaining the formation. We address this problem for a class of dynamic unicycle robots. Using Inputto-State Stability we combine a general class of formationkeepi...

We present a collision avoidance system for powered wheelchairs used by people with cognitive disabilities. Such systems increase mobility and feelings of independence, thereby enabling reversal of some symptoms of depression and cognitive impairment and improvement of quality of life. We use a novel 3D sensor developed by Canesta Inc. that allows the wheelchair to “see” obstacles, avoid collis...

Obstacles are represented in the attentional landscape. However, it is currently unclear what the exclusive contribution of attention is to the avoidance response. This is because in earlier obstacle avoidance designs, it was impossible to disentangle an effect of attention from the changing features of the obstacle (e.g., its identity, size, or orientation). Conversely, any feature manipulatio...