Coexistence of humans and mobile robots

Pedestrian movement is considered to be influenced by the environment, the pedestrian's own intentions, and other pedestrians. Taking these effects into account, we aim to accurately predict the movement of pedestrians.

Prediction of pedestrian movement considering the relationship with others in the vicinity

Pedestrians move under the influence of others around them, and the degree of this influence depends on the relationship between the pedestrians. We propose a method for learning such interactions with others in the vicinity using machine learning approaches to predict pedestrians.

  • Jiaxu Wu, Hanwool Woo, Yusuke Tamura, Alessandro Moro, Stefano Massaroli, Atsushi Yamashita, and Hajime Asama: Pedestrian Trajectory Prediction Using BiRNN Encoder-Decoder Framework, Advanced Robotics, Vol.33, No.18, pp.956-959, 2019. doi:10.1080/01691864.2019.1635910

Detection of "smartphone zombies" from a mobile robot

With the spread of smartphones in recent years, smartphone zombies have become a major social problem. When someone's attention is focused on his/her smartphone, he/she lose attention to his/her surroundings and run the risk of colliding with others. We have proposed a method to detect such high-risk pedestrians by using a LiDAR mounted on a mobile robot.

  • Jiaxu Wu, Yusuke Tamura, Yusheng Wang, Hanwool Woo, Alessandro Moro, Atsushi Yamashita, and Hajime Asama: Smartphone Zombie Detection from LiDAR Point Cloud for Mobile Robot Safety, IEEE Robotics and Automation Letters, Vol.5, No.2, pp.2256-2263, 2020. doi:10.1109/LRA.2020.2970570

Prediction of pedestrian movement based on environmental information

For example, when walking down a straight corridor, a human walks in an approximate straight line along the direction of the corridor. In this way, humans move in a way that is influenced by the shape of their environment. Also, when you walk in an office, for example, the meaning of the environment is greatly influenced by what rooms are in the vicinity, where the elevators are, where the restrooms are, and so on. Therefore, it is important to consider this environmental information in order to accurately predict the movement of pedestrians.
We have proposed a method to accumulate the movement characteristics of pedestrians in the environment by using sensor information installed in the environment, and use it for movement prediction.

  • Shunsuke Hamasaki, Yusuke Tamura, Atsushi Yamashita and Hajime Asama: Prediction of Human’s Movement for Collision Avoidance of Mobile Robot, Proceedings of the 2011 IEEE International Conference on Robotics and Biomimetics, pp.1633-1638, Phuket, Thailand, December 2011. doi:10.1109/ROBIO.2011.6181523

Pedestrian behavior model considering intention

Humans are thought to be walking while switching their intentions, such as avoiding or following others in the vicinity, depending on the situation of others and the environment around them. We extend the Social Force Model1 and propose a model that generates behaviors while generating sub-goals according to the switching of such walking intentions.

  • Yusuke Tamura, Phuoc Dai Le, Kentarou Hitomi, Naiwala P. Chandrasiri, Takashi Bando, Atsushi Yamashita and Hajime Asama: Development of Pedestrian Behavior Model Taking Account of Intention, Proceedings of the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp.382-387, Vilamoura, Portugal, October 2012. doi:10.1109/IROS.2012.6385599

  1. A model in which pedestrians are regarded as a point mass and move according to virtual forces from the environment, others, and destinations acting on them. (Helbing & Molnár, 1995)。