Jerry Yao | Bike guards

Having the bike sharing system as your bodyguard

Ever wandering in the city alone late at night feeling insecure? What if we can "hire" the public bike-sharing system as our bodyguard? City Bike Guard is a function on PEV (lightweight autonomous bike), it follows behind the user quietly while showing projection on the floor, visualizing the world that robot perceives and applying it to the real world to raise the pedestrians' sense of security in the city.

The embedded vision system records footage when emergencies occur and upload the footage to cloud, as if the pedestrians are being watched and guarded. It re-thinks the way we interact with mobility system and provides a new mean for pedestrians to bound with the city.

My role

As a design engineer in this project. I design mechanical system, human-machine interface (HMI), and projection mapping technique. I built this prototype and work with software engineers to extract data from the LiDAR and visualize it.

Skills & software

Prototyping, 3D modeling, machining, interaction design.
Rhinoceros, Laser Cut, Water jet, ROS, Python, Javascript.

Team

Justin Zhang, Luke Jiang.

Learn more from the Media Lab project website.

Problem

How to establish trust between PEV and the users?

New modes of 21st century urban transportation are becoming increasingly lightweight, electrified, connected, shared, and autonomous.

Cohabitation of humans and machines is an increasingly important question, and one which requires careful attention and design. We strive to enable new forms of human-machine co-existence, trust, and collaboration.

Proposed solution

Two-way communication

Intuitive and effective two-way communication between vehicles and pedestrians, facilitating coexistence, trust-building, and collaboration among people and machines.

Regulations

Street safety and traffic-yielding mechanisms.

Learn from bikes

Behavior change related to the adoption of active mobility mode, or electric assist.

Design & prototype

Research

Human Robot Interaction (Jibo), HMI research (Audi Aicon), Mercedes-Benz F 015 (projection).

Design

Build mechanical 3d model, design projection scenarios, hardware system design.

Prototype

Laser cut, water jet, hardware assemble, projection mapping, openEV, pose detection interactive prototype.

Integration

Collaborate with software teams and link the design system to ROS.

Projection mapping and LiDAR fusion

The LiDAR data that shown in the video is how PEV (a robot) perceives the world. The inflated blue circles represents the obstacles. We extract the center points of the obstacles from LiDAR's data and visualize it with projection, mapping with the obstacle in the real world. It also works with multi users.

Using projection to show the intention of an autonomous vehicle

The Projection could also be used to express the intention of the vehicle. In the future we will develop more personalized projection and use it to express the emotions of the vehicle.

Using human posture to build interaction between human and the vehicle.

The vision system on PEV also enables pedestrian to use posture as an input to interact with PEV.

Result and lessons learned

This project was showcased at the 2018 MIT Media Lab Member's event and received positive feedback from out sponsor company Denso Corporation. Denso will keep working with us on the development of Human Machine Interaction for PEV.

This is my first time working on projection mapping project. There is still a lot to improve, yet I do see the potential of projection being applied to more scenarios. For instance, if PEV was an agent in the theme park to help carry visitors' belongings while walking in the park. The projection would be a really intuitive way for people to interact with. We can also design more personalized projects that relate to the theme park and attract visitors to communicate with the robot.

City bike guards