Detecting obstacle distance of horizontal 360 degrees range is needed in robotics navigation and obstacle avoidance algorithms.The conventional distance finding sensors include laser, ultrasonic, infrared, microwave, and so on.Now distance extraction based on computer vision is attractive in robotics.Video cameras provide a large amount of data at high speed and low cost.The purpose of robot vision is to enable robots to perceive the external world in order to perform a large range of tasks such as navigation, visual servoing for object tracking and manipulation, object recognition and categorization, surveillance, and higher-level decisionmaking.Distance measurement in computer vision and robotics is commonly done via stereo (binocular) vision and monocular vision.In binocular vision, images from two cameras are used to triangulate and estimate distances.The basic procedure includes camera calibration,metrical model, feature extraction, stereo matching.Monocular cues provide depth information though motion parallax, kinetic depth perception, perspective, relative size, familiar size, aerial perspective, occlusion, peripheral vision, and texture gradient.In recent years, machine learning has been studied widely and combined with computer vision.From the standpoint of computer vision systems, machine learning can offer effective methods for automating the acquisition of visual models, adapting task parameters and representation, transforming signals to symbols,building trainable image processing systems, focusing attention on target object.There are many research literatures about depth estimation from single monocular images using supervised learning approach.This paper presents a distance measurement project using a multi-lens panoramic camera (e.g.Ladybug3).The camera has six CCD sensors, five at the side and one on the top, and can take 360 degrees panoramic image.The five lenses at the side can be used respectively.The project proposed herein is the integration of binocular vision method, monocular vision, and machine learning approach.Under the condition that the lens, which have certain shooting angles, are stationary, binocular vision distance measurement method is used in the overlapping areas of adjacent lens.In the other areas, monocular vision together with machine learning approach is used to acquire the depth map and then obtain the distance data.Point Grey Ladybug SDK is used in camera calibration, images acquisition.MATLAB is used in image processing.The measurement result of obstacle distance of horizontal 360 degrees range can be used in robotics navigation and obstacle avoidance algorithms (e.g.BUG).More robotics vision application based on the achievement in this paper will be studied in the future.