Thomas Lips

Thomas Lips, Victor-Louis De Gusseme, Francis wyffels

Robotic cloth manipulation is challenging due to its deformability, which makes determining its full state infeasible. However, for cloth folding, it suffices to know the position of a few semantic keypoints. Convolutional neural networks (CNN) can be used to detect these keypoints, but require large amounts of annotated data, which is expensive to collect. To overcome this, we propose to learn these keypoint detectors purely from synthetic data, enabling low-cost data collection. In this paper, we procedurally generate images of towels and use them to train a CNN. We evaluate the performance of this detector for folding towels on a unimanual robot setup and find that the grasp and fold success rates are 77% and 53%, respectively. We conclude that learning keypoint detectors from synthetic data for cloth folding and related tasks is a promising research direction, discuss some failures and relate them to future work. A video of the system, as well as the codebase, more details on the CNN architecture and the training setup can be found at https://github.com/tlpss/workshop-icra-2022-cloth-keypoints.git.

Thomas Lips, Marco Moletta, Michael C. Welle et al.

RGB-based imitation learning requires many demonstrations to generalize to unseen objects or scenes, motivating research into intermediate representations to improve generalization for robotic manipulation. Visual foundation models enable one-shot extraction of keypoints to provide such representation. However, it remains unclear how to integrate them into imitation learning optimally and when they outperform alternative representations. We combine approaches from previous works on keypoint imitation learning (KIL) and investigate several design choices to provide practical guidelines. Using over 2000 real-world rollouts, we also assess the generalization capabilities of KIL to unseen objects and scene variations. KIL achieves a 75% overall success rate across five tasks, significantly outperforming the RGB baseline (47%) and performing on par with S2-diffusion (73%). Finally, we explore the limitations of the foundation models used for keypoint extraction and extend KIL to tasks with multiple object instances. Our results confirm KIL as a data-efficient approach for robot learning, though it does not outperform alternative representations and inherits limitations of the foundation models used for keypoint extraction. All rollout videos, demonstrations, and results are available at https://kil-manipulation.github.io/.

Remko Proesmans, Thomas Lips, Francis wyffels

Large behaviour models have transformed the field of robotic manipulation, but prohibitive data requirements have thus far prevented a revolution similar to vision language models. We believe that instrumentation, i.e. sensor integration in objects, can provide invaluable state information and enable efficient learning for robotic manipulation. In this paper, we present instrumented imitation learning of clothes hanger insertion. Using 180 teleoperated demonstrations, we train diffusion policies with and without access to instrumentation data. Results show that policies leveraging instrumentation outperform vision-only counterparts by 14-25 %pt and exhibit greater task awareness. Crucially, a black-box imitation learning policy learns to prioritise instrumentation signals without explicit guidance. In addition, enhancing the teleoperation dataset with rollouts from an instrumented expert policy, enables a vision-only student policy to achieve performance comparable to the instrumented expert, thereby surpassing the original vision-only policy. These findings establish instrumentation as a promising strategy to enhance imitation learning for robotic manipulation. Datasets are available on Zenodo.

Thomas Lips, Victor-Louis De Gusseme, Francis wyffels

Assistive robots should be able to wash, fold or iron clothes. However, due to the variety, deformability and self-occlusions of clothes, creating robot systems for cloth manipulation is challenging. Synthetic data is a promising direction to improve generalization, but the sim-to-real gap limits its effectiveness. To advance the use of synthetic data for cloth manipulation tasks such as robotic folding, we present a synthetic data pipeline to train keypoint detectors for almost-flattened cloth items. To evaluate its performance, we have also collected a real-world dataset. We train detectors for both T-shirts, towels and shorts and obtain an average precision of 64% and an average keypoint distance of 18 pixels. Fine-tuning on real-world data improves performance to 74% mAP and an average distance of only 9 pixels. Furthermore, we describe failure modes of the keypoint detectors and compare different approaches to obtain cloth meshes and materials. We also quantify the remaining sim-to-real gap and argue that further improvements to the fidelity of cloth assets will be required to further reduce this gap. The code, dataset and trained models are available

Raman Talwar, Remko Proesmans, Thomas Lips et al.

Learning contact-rich manipulation is difficult from cameras and proprioception alone because contact events are only partially observed. We test whether training-time instrumentation, i.e., object sensorisation, can improve policy performance without creating deployment-time dependencies. Specifically, we study button pressing as a testbed and use a microphone fingertip to capture contact-relevant audio. We use an instrumented button-state signal as privileged supervision to fine-tune an audio encoder into a contact event detector. We combine the resulting representation with imitation learning using three strategies, such that the policy only uses vision and audio during inference. Button press success rates are similar across methods, but instrumentation-guided audio representations consistently reduce contact force. These results support instrumentation as a practical training-time auxiliary objective for learning contact-rich manipulation policies.

Thomas Lips, Francis wyffels

Robots that assist humans will need to interact with articulated objects such as cabinets or microwaves. Early work on creating systems for doing so used proprioceptive sensing to estimate joint mechanisms during contact. However, nowadays, almost all systems use only vision and no longer consider proprioceptive information during contact. We believe that proprioceptive information during contact is a valuable source of information and did not find clear motivation for not using it in the literature. Therefore, in this paper, we create a system that, starting from a given grasp, uses proprioceptive sensing to open cabinets with a position-controlled robot and a parallel gripper. We perform a qualitative evaluation of this system, where we find that slip between the gripper and handle limits the performance. Nonetheless, we find that the system already performs quite well. This poses the question: should we make more use of proprioceptive information during contact in articulated object manipulation systems, or is it not worth the added complexity, and can we manage with vision alone? We do not have an answer to this question, but we hope to spark some discussion on the matter. The codebase and videos of the system are available at https://tlpss.github.io/revisiting-proprioception-for-articulated-manipulation/.