Lio - A Personal Robot Assistant for Human-Robot Interaction and Care Applications | IROS 2020
Justin Smith, F&P Robotics’ AI lead, presents Lio’s design and software for human-robot interaction and care applications at IROS 2020; colleagues from software development and care robotics also describe its decision system and deployments.
In this talk
- 0:05
Safety and physical design
Smith introduces Lio as a personal robot built for interaction and healthcare. Its mobile base uses two LiDAR sensors, bumpers, and floor sensors to navigate and stop at obstacles or drops. A perimeter light communicates travel direction and can signal faults; the padded arm can yield to a person’s touch, while collision detection stops motion after contact.
- 3:03
Adaptable arm and gripper
An interactive ring provides controls for releasing the arm, saving poses, and other fixed or configurable tasks. The gripper uses a camera and proximity sensors in its fingers and palm to estimate distance and detect whether an object was grasped or dropped. Fingers and gripper heads can be replaced, including with a project-specific 3D-printed finger.
- 6:18
Interfaces and programming
Lio offers separate web interfaces for home users, nursing staff, and ROS module monitoring, plus MyP for writing and deploying Python behaviors. Its simulator stays synchronized with the physical robot, and users can build motions from saved poses, edit paths, and check trajectories before running them on the robot. Mapping tools support facility maps, waypoints, virtual obstacles, and no-go zones.
- 9:25
Autonomy and care uses
Fedex Filling describes a Prolog-based decision engine that weighs environmental facts and rules to select actions, such as charging when the battery falls below a threshold and driving is permitted. A priority step considers user inputs and scheduled tasks before execution. Lio also uses skeleton tracking to detect people and estimate poses, and Michael says it is used at healthcare and research institutions across Europe for activities including music, VR therapy, transport, physical support, security, and disinfection.
- 13:21
Demonstration and transport
Lio demonstrates delivering a drink and prompting the recipient to take a sip. Michael explains that the robot folds into a compact position for transport in a regular car, allowing teams to move it between facilities and set it up again.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.