2A2-Q02 Personal Robot to Support Active Human Life by Assisting Transportation : Person detection and following(Robots for Home/Office Application)

Abstract

This paper describes a novel personal robot to support active human life by assisting transportation. The proposed robot is able to follow a person with bringing the luggage. Therefore, the robot encourages the elder people to go to outside for shopping, visiting friends and so on. In order to realize these requested specifications, the robot has to achieve both small footprint for coexistence with people and high traveling performance to follow a person.In this paper, the method for tracking and following a person are proposed to realize robust person following at speed as ordinary walkers. The proposed tracking method based on extended Kalman filter is able to track peoples around the personal robot using the data from laser range finder and the robot velocity as inputs. The following controller is based on direct following. The effectiveness of the proposed methods is verified through the experiment using the new prototype robot.

Personal robot assisting transportation to support active human life — Human-following method based on model predictive control for adjacency without collision

“…Assumption Figure 9 shows the top view when the personal robot follows the human being. Here, x h , y h represents the coordinates of the human being, x r , y r represents the coordinates of the personal robot, and σ rh , d rh , and v h can be estimated by the extended Kalman filter [10], [11]. C hf has to generate , which must be larger than zero for the no collision.…”

“…The roll angle of the upper body can be controlled by the posture actuator, which has the same composition as that in the third prototype robot. A laser range finder and an omnidirectional camera are installed at the top of the robot to detect and track the human being [10], [11]. Also, an IMU (inertial measurement unit) is mounted to control the roll angle and to realize wheeled inverted pendulum control.…”

“…The predicted value X ha of x h for case (a) and the predicted value X hb of x h for case (b) can be given as follows: (11) where Σ x h , Γ xh , and Θ xh are the matrix coefficients by [13], and v max is the assumed maximum velocity of the human being. By combining (4), (5), (10), and (11), the predicted relative distance D rh and the predicted pitch angle Φ for cases (a) and (b) can be given as follows:…”

Posture Stabilization Control of Personal Robot Based on Model Predictive Control

“…The method for detecting the owner by using these sensors is described in Refs. [7] and [8], and so is omitted in this paper. C wip : the feedback control rule representing inverted twowheel control; C gr : the feedback control rule that controls the turning angular velocity; C r : the feedback control rule that controls the roll angle; τ v : the control input that C wip is calculated from; τ ω : the control input that C yr is calculated from; and Q: the matrix given below:…”

“…The specific algorithm for C hf can be found in Refs. [4] and [7]. C hf gives as control input the forward speed reference v ref and the turning angular velocity reference ω refc .…”