Our highlights from IROS 2022
We were fortunate to be able to contribute to the IROS 2022 Conference – the IEEE/RSJ International Conference on Intelligent Robots Systems in Japan this year. The annual IROS event aims to create opportunities to meet, discuss, and become familiar with the most recent advances in the field of robotics, autonomous and intelligent systems, human-compliant robots, medical robotics, and more. Read on to find out more about our highlights of the event.
IROS has the goal of offering a place to gather new findings and ideas, with the opportunity to discuss them across the robotics community. The event is sponsored by IEEE (Institute of Electrical and Electronics Engineers), RSJ (The Robotics Society of Japan), IEEE Robotics & Automation Society, IES (the IEEE Industrial Electronics Society), SICE (Society of Instrument and Control Engineers), and the NIF (New Technology Foundation).
We had the opportunity to actively participate in a number of talks and workshops, as well as have our own booth together with our TIAGo robot. Here we share more on our participation in this key annual event in the robotics calendar:
Tutorial on ROS4HRI – From Zero to Multi-Modal Interactive Dialogue for Robots
Our colleagues Séverin Lemaignan and Lorenzo Ferrini hosted the workshop “From Zero to Multi-Modal Interactive Dialogue for Robots” on ROS for HRI, together with Youssef Mohamed of KTH in Stockholm,
In this workshop, they presented the new ‘ROS4HRI’ ROS standard for human-robot interaction and programmed a small simulated interactive system able to track and react to faces, skeletons, and chats. The attendees at the workshop learned about ROS4HRI specifications, as well as gaining hands-on experience with several key open-source, ROS4HRI-compatible, nodes for Human-Robot Interaction (including MediaPipe-based face detection, skeleton tracking, speech recognition, and dialogue management).
ROS is widely used in the context of human-robot interactions (HRI). However, to date, there has been limited success in coming up with broadly accepted interfaces and pipelines for that domain, as found in other parts of the ROS ecosystem. The ROS4HRI framework has been recently introduced to address this issue. The ROS4HRI framework is available on PAL Robotics’ ARI robot, and you can find out more on the Wiki ROS. Additionally, we have a complete walk through tutorial on ROS4HRI available that users can try for themselves at home. To ask more about ROS4HRI don’t hesitate to get in touch with our experts.
Workshop on Mobile Manipulation and Embodied Intelligence (MOMA): Challenges and Opportunities
PAL Robotics was a sponsor at the MOMA workshop, which aimed to address the ongoing challenges and opportunities of researching embodied AI mobile manipulator robots in real-world unstructured environments, bringing together a broader community of researchers, from robotics, AI planning, computer vision, and machine learning, to discuss and contribute to the challenging topic of embodied AI mobile manipulation.
The workshops’ objectives included discussing some of the following research questions:
- How can we effectively coordinate the properties (embodiments) of mobile manipulators?
- How can we encode (represent) the multimodal information of dynamically changing environments for planning mobile manipulation tasks?
- How can we effectively integrate learning for mobile manipulation? What are the challenges and benefits?
- How can high-level decision-making affect low-level mobile manipulation execution?
Concluding this session, an awards ceremony was held, for which PAL Robotics’ CTO Luca Marchionni presented Daniel Honerkamp with the best paper award for his work on Neural Navigation for Mobile Manipulation (see more details below). This workshop was organised by Georgia Chalvatzaki (TU Darmstadt), Roberto Martin-Martin (University of Texas), Nikolaus Correll (University of Colorado Boulder), Kensuke Harada (Osaka University), Tamim Asfour (KIT), Mehmet Dogar (University of Leeds).
MOMA workshop: research with PAL Robotics’ TIAGo the mobile manipulator
Included in Workshop on Mobile Manipulation and Embodied Intelligence (MOMA) were a number of research papers that featured our TIAGo robot. For example:
Robot Learning of Mobile Manipulation with Reachability Behavior Priors by Snehal Jauhri, Jan Peters, and Georgia Chalvatzaki – this work studies the integration of robotic reachability priors in actor-critic RL methods for accelerating the learning of MM for reaching and fetching tasks. TIAGo++ took part in this research which was the winner of the ‘Mobile Manipulation best paper award.’
Why did I fail? A Causal-based Method to Find Explanations for Robot Failures by Maximilian Diehl and Karinne Ramirez-Amaro, this paper proposes a novel method that enables robots to generate contrastive explanations upon task failures. One of the robotic platforms they ran their experiments on was the TIAGo mobile manipulator with one arm and a parallel gripper.
N2M2: Learning Navigation for Arbitrary Mobile Manipulation Motions in Unseen and Dynamic Environments by Daniel Honerkamp, Tim Welschehold, and Abhinav Valada introduced Neural Navigation for Mobile Manipulation (N2M2) which extends this decomposition to complex obstacle environments and enables it to tackle a broad range of tasks in real world settings. TIAGo with a differential drive, height-adjustable torso and a flexible 7-DOF arm took part in this research.
Talk on research paper “Bio-Inspired Grasping Controller for Sensorized 2-DoF Grippers”
Our colleague Luca Lach delivered a talk on his recent research paper, “Bio-Inspired Grasping Controller for Sensorized 2-DoF Grippers.” The research presents a holistic grasping controller, combining free-space position control and in-contact force-control for reliable grasping given uncertain object pose estimates.
The controller was evaluated in two experiments on PAL Robotics’ TIAGo mobile manipulator robot and its parallel-jaw gripper proving the effectiveness of the approach for robust grasping and minimising object displacement.
The authors of this paper were: Luca Lach (PAL Robotics), Séverin Lemaignan (PAL Robotics), Francesco Ferro (PAL Robotics), Helge Ritter (Bielefeld University), and Robert Haschke (Bielefeld University).
Workshop on Social and Cognitive Interactions for Assistive Robotics (SCIAR)
Our Senior Scientist in social robotics, Séverin Lemaignan co-organised a workshop on “Social and Cognitive Interactions for Assistive Robotics (SCIAR)”. This workshop aimed to develop computational approaches for natural and social human-robot interaction for Assistive Robotics Technology. Attention was also given to the cognitive interaction analysis that features the interpretation of human interactions, with a particular emphasis on the nonverbal signals, in order to facilitate a more natural and social interaction.
The programme included keynotes on topics such as, “Social robots for assessing and promoting mental well-being,” “Personalising interactions with assistive robots,” and, “Robots that understand us.”
This workshop was organised by Cigdem Beyan (University of Trento), Yiming Wang (FBK), Fei Chen (Chinese University of Hong Kong), Séverin Lemaignan (PAL Robotics), and Elisa Ricci (University of Treno/FBK).
Talk on research paper “Deliberative Democracy with swarm robots”
PAL Robotics’ Séverin Lemaignan presented the research paper, “Deliberative Democracy with swarm robots” together with Merihan Alhafnawi, Robert Hunt, Paul J O’Dowd, and Sabine Hauert (all Bristol University).
This work tests whether a swarm of robots can help facilitate decision-making by visually representing the diversity of opinions. They used a swarm of robots they built, that consisted of 4-inch touchscreens-on-wheels. The robots acted as physical avatars for opinions, helping them travel and mix together. Their contribution was in testing the possibility of how and whether using a moving robot swarm helps humans reach a decision.
PAL Robotics booth and TIAGo the mobile manipulator robot
PAL Robotics attended IROS, together with our team of robotics experts and our TIAGo the mobile manipulator robot. TIAGo combines perception, navigation, manipulation & Human-Robot Interaction skills out of the box, and is one of the most used robots for research in universities and innovation centers worldwide.
PAL Robotics’ TIAGo robot is also a chosen platform for robotics competitions due to the robot’s modular design, flexibility, and the option for teams to access the platform in simulation. Find out more about the robotics competitions at IROS 2022.
EU and collaborative projects in agrifood, healthcare, personal robots and more
We take part in many collaborative and EU projects, which we also talked to visitors about at the event. Here are some of our current projects that were featured at this year’s IROS:
- CANOPIES has the goal of developing a Human-Robot prototype in crop farming (Agri-Food) within a table grape vineyard.
- NeuTouch aims at improving artificial tactile systems, and developing a new type of technology that is based on the same principles.
- SPRING is working on the adaptation of Socially Assistive Robots, with the ability to perform multi-person interactions in a hospital environment.
- TALBOT will develop a cognitive architecture for humanoid robots supporting adaptability and validate the proposed technology in healthcare applicative scenarios.
- OpenDR aims to harness deep learning to provide advanced perception and cognition abilities in robotics and use these robotic applications in areas such as healthcare, agri-food and agile production.
- PERSEO aims to train a new generation of research and professional figures to face research challenges in the market of personal robots.
- Evolved-5G aims to contribute towards the fourth industrial revolution or Industry 4.0, and during the project NetApp will be tested, validated, and certified.
- SHAPES aims to create an environment that supports deployment of digital solutions to support healthy and independent living for older individuals.
We thoroughly enjoyed participating in the IROS 2022 Conference in Japan and we would like to thank all of the organisers and all those that attended our sessions, for what was once again a great event. We look forward to IROS 2023! To find out more about PAL Robotics and our robots, visit our website. To talk with us about collaborating or to ask any questions, don’t hesitate to get in touch.