Articles worth reading
Sero Samsung’s Smart QLED 4K UHD HDR TV

Sero Samsung’s Smart QLED 4K UHD HDR TV

access_time August 1, 2020

Samsung, the South Korean electronics giant, launched Sero, its world’s first mobile-friendly Smart TV with a rotating screen to cast

NCAA: How We Are Using Predictive Analytics during the Final Four

NCAA: How We Are Using Predictive Analytics during the Final Four

access_time April 3, 2018

NCAA is linked to Google Cloud Google’s cloud team is experimenting with large amounts of data and real-time analysis and

Google Bans Artificial Intelligence for Weapon Use

Google Bans Artificial Intelligence for Weapon Use

access_time June 26, 2018

With Artificial Intelligence for Weapons on Moral Grounds According to the Google leadership, the company will not renew its Project

Blog

New Robot Rolls with the Rules of Pedestrian Conduct

New Robot Rolls with the Rules of Pedestrian Conduct

access_time July 29, 2017 remove_red_eye 6087 views

 

Robot rolls with the rules of pedestrian conduct

In the same way that drivers follow the traffic code, most pedestrians follow certain social codes when it comes to moving through a crowded public thoroughfare: staying on the right, overtaking on the left, maintaining a respectable distance, and be prepared to deviate or change trajectory in order to avoid obstacles, while maintaining a steady pace of progress.

Now, engineers at the Massachusetts Institute of Technology (MIT) in Cambridge, United States, have designed an autonomous robot with “socially conscious navigation” that can travel between pedestrians while respecting the general codes of conduct of the pedestrians The system of locomotive behavior used in this robot could allow other robots to move through hospitals, shopping malls and other areas with a lot of pedestrian traffic.

In tests carried out at MIT facilities, the robot was able to avoid collisions while maintaining adequate speed in its advance, in line with the average pedestrian flow.

The investigation is the work of Yu Fan team “Steven” Chen, Michael Everett, Miao Liu and Jonathan How.

For a robot to be able to move autonomously through an environment with a lot of pedestrian or vehicle traffic, it must solve four main challenges: location (knowing exactly where it is located), perception (recognizing its surroundings), movement planning optimum to a given destination) and control (physically perform the desired path).

For perception, Chen and his colleagues equipped the robot with sensors already available, such as webcams, a depth sensor and a high resolution LIDAR sensor. For localization, they used open-source algorithms in order to map the robot environment and determine its position. To control it, they used models already used for autonomous land vehicles.

folder_openAssigned tags
content_copyCategorized under