Detecting human actions is a crucial task for autonomous robots and vehicles, often requiring the integration of various data modalities for improved accuracy. In this study, we introduce a novel approach to Human Action Recognition (HAR) using language supervision named LS-HAR based on skeleton and visual cues. Our method leverages a language model to guide the feature extraction process in the skeleton encoder. Specifically, we employ learnable prompts for the language model conditioned on the skeleton modality to optimize feature representation. Furthermore, we propose a fusion mechanism that combines dual-modality features using a salient fusion module, incorporating attention and transformer mechanisms to address the modalities’ high dimensionality. This fusion process prioritizes informative video frames and body joints, enhancing the recognition accuracy of human actions. Additionally, we introduce a new dataset tailored for real-world robotic applications in construction sites, featuring visual, skeleton, and depth data modalities, named VolvoConstAct. This dataset serves to facilitate the training and evaluation of machine learning models to instruct autonomous construction machines for performing necessary tasks in real-world construction sites. To evaluate our approach, we conduct experiments on our dataset as well as three widely used public datasets: NTU-RGB+D, NTU-RGB+D 120, and NW-UCLA. Results reveal that our proposed method achieves promising performance across all datasets, demonstrating its robustness and potential for various applications. The code, dataset, and demonstration of real-machine experiments are available at: https://mmahdavian.github.io/ls_har/

In end-to-end autonomous driving, current sensor fusion and navigational control techniques used by imitation learning algorithms are insufficient in challenging scenarios involving multiple dynamic agents and result in poor driving capabilities. To tackle this issue, we introduce DMFuser, a transformer-based algorithm that employs knowledge distillation between multi-task student and single-task teachers and combines attention and convolutions to fuse multiple RGB-D camera representations to produce vehicular navigational commands (throttle, steering and brake). Our model incorporates two modules. The first module, perception, encodes data from RGB-D cameras for tasks like semantic segmentation, semantic depth cloud (SDC) mapping, and traffic light state recognition. To enhance feature extraction and fusion from both RGB and depth sources, we harness local and global capabilities of convolution and transformer modules. We employ an attention-CNN fusion structure to effectively learn and fuse RGB and SDC map features. Subsequently, the control module decodes these features along with supplementary data, containing environment’s static and dynamic information, to predict waypoints and vehicular control actions. We evaluate the model and conduct a comparative analysis, in various scenarios, weather conditions, and traffic situations, spanning from normal to adversarial in the CARLA simulator. We achieve better or comparable results in term of driving score (DS) and other metrics with respect to our baselines. Also, our ablation studies demonstrate the effectiveness of our contributions to improve the driving skills. Our code is available at the following github page: https://github.com/pagand/e2etransfuser

In this paper, we greatly expand the capability of robots to perform the follow-ahead task and variations of this task through development of a neural network model to predict future human motion from an observed human motion history. We propose a non-autoregressive transformer architecture to leverage its parallel nature for easier training and fast, accurate predictions at test time. The proposed architecture divides human motion prediction into two parts: 1) the human trajectory, which is the 3D positions of the hip joint over time, and 2) the human pose which is the 3D positions of all other joints over time with respect to a fixed hip joint. We propose to make the two predictions simultaneously, as the shared representation can improve the model performance. Therefore, the model consists of two sets of encoders and decoders. First, a multi-head attention module applied to encoder outputs improves human trajectory. Second, another multi-head self-attention module applied to encoder outputs concatenated with decoder outputs facilitates the learning of temporal dependencies. Our model is well-suited for robotic applications in terms of test accuracy and speed, and compares favorably with respect to state-of-the-art methods. We demonstrate the real-world applicability of our work via the Robot Follow-Ahead task, a challenging yet practical case study for our proposed model. The human motion predicted by our model enables the robot follow-ahead in scenarios that require taking detailed human motion into account such as sit-to-stand, stand-to-sit. It also enables simple control policies to trivially generalize to many different variations of human following, such as follow-beside. Our code and data are available at the following Github page: https://github.com/mmahdavian/STPOTR

Human body motion prediction is a fundamental part of many human-robot applications. Despite the recent progress in the area, most studies predict human body motion relative to a fixed joint and only limit their model to predict one possible future motion, or both. However, due to the complex nature of human motion, a single prediction cannot adequately reflect the many possible movements one can make. Also, for any robotics application, prediction of the full human body motion including the absolute 3D trajectory - not just a 3D body pose relative to the hip joint - is needed. In this paper, we try to address these two shortcomings by proposing a transformer-based generative model for forecasting multiple diverse human motions. Our model generates$N$future possible body motions given the human motion history. This is achieved by first predicting the pose of the body relative to the hip joint as was done in prior work. Then, our proposed Hip Prediction Module predicts the trajectory of the hip position relative to a global reference frame for each predicted pose frame, an aspect of human body motion neglected by previous work. To obtain a set of diverse predicted motions, we introduce a similarity loss that penalizes the pairwise sample distance. Our system not only outperforms the state-of-the-art in human motion prediction, but also is able to predict a diverse set of future human body motions, including the hip trajectory.