IROS 2020: Autonomous mail delivery, robots practicing bartending, and more
San Diego, Calif., Nov. 8, 2020 -- From autonomous vehicles to robots practicing bartending and insect-like robots, engineers at the University of California San Diego are showcasing a broad range of pacers at IROS 2020, which is being held virtually from Oct. 25 to Nov. 25.
The Contextual Robotics Institute at UC San Diego is a full stack research enterprise, from autonomy, to robots in medicine, to human robot interaction, said institute director Henrik Christensen, who is also a professor in the UC SanDiego Department of Computer Science.
“The conference is a unique opportunity for the Contextual Robotics Institute to showcase our diverse research portfolio and to engage with the broader audience to demonstrate how robots are changing the world from manufacturing to e-commerce to helping in everyday life,” Christensen said. “IROS is also for the first time putting a significant emphasis on diversity equity and inclusion, which is a great new direction.”
For the first time, the conference is available for free after registration. This year’s theme is Consumer Robotics and Our Future.
Laurel Riek, a professor in the UC San Diego Department of Computer Science and Engineering, is giving an invited talk at the RoPat20 workshop. She is speaking on “Expressive Patient Simulators for Clinical Education.”
Nicholas Gravish, a professor in the Department of Mechanical and Aerospace Engineering at UC San Diego, is one of the organizers of a workshop on robotics-inspired biology. In addition, Michael Yip, a professor in the Department of Electrical and Computer Engineering, and one of his PhD students, Florian Richter, are co-organizers of a workshop on cognitive robotics surgery.
Here is a listing of the papers that UC San Diego faculty are contributing to the conference this year:
Autonomous Vehicle Benchmarking using Unbiased Metrics
David Paz, Po-jung Lai, Nathan Chan, Yuqing Jiang and Henrik I. Christensen
With the recent development of autonomous vehi- cle technology, there have been active efforts on the deployment of this technology at different scales that include urban and highway driving. While many of the prototypes showcased have been shown to operate under specific cases, little effort has been made to better understand their shortcomings and generalizability to new areas. Distance, uptime and number of manual disengagements performed during autonomous driving provide a high-level idea on the performance of an autonomous system but without proper data normalization, testing location information, and the number of vehicles involved in testing, the disengagement reports alone do not fully encompass system performance and robustness. Thus, in this study a complete set of metrics are applied for benchmarking autonomous vehicle systems in a variety of scenarios that can be extended for comparison with human drivers and other autonomous vehicle systems. These metrics have been used to benchmark UC San Diego’s autonomous vehicle platforms during early deployments for micro-transit and autonomous mail delivery applications.
Probabilistic Semantic Mapping for Urban Autonomous Driving Applications
David Paz, Hengyuan Zhang, Qinru Li, Hao Xiang and Henrik I. Christensen
Recent advancements in statistical learning and computational abilities have enabled autonomous vehicle tech- nology to develop at a much faster rate. While many of the architectures previously introduced are capable of op- erating under highly dynamic environments, many of these are constrained to smaller-scale deployments, require constant maintenance due to the associated scalability cost with high- definition (HD) maps, and involve tedious manual labeling. As an attempt to tackle this problem, we propose to fuse image and pre-built point cloud map information to perform automatic and accurate labeling of static landmarks such as roads, sidewalks, crosswalks, and lanes. The method performs semantic segmentation on 2D images, associates the semantic labels with point cloud maps to accurately localize them in the world, and leverages the confusion matrix formulation to construct a probabilistic semantic map in bird’s eye view from semantic point clouds. Experiments from data collected in an urban environment show that this model is able to predict most road features and can be extended for automatically incorporating road features into HD maps with potential future work directions.
Neural Manipulation Planning on Constraint Manifolds
Ahmed H. Qureshi, Jiangeng Dong, Austin Choe, and Michael Yip
The presence of task constraints imposes a significant challenge to motion planning. Despite all recent advancements, existing algorithms are still computationally expensive for most planning problems. In this paper, we present Constrained Motion Planning Networks (CoMPNet), the first neural planner for multimodal kinematic constraints. Our approach comprises the following components: i) constraint and environment perception encoders; ii) neural robot configuration generator that outputs configurations on/near the constraint manifold(s), and iii) a bidirectional planning algorithm that takes the generated configurations to create a feasible robot motion trajectory. We show that CoMPNet solves practical motion planning tasks involving both unconstrained and constrained problems. Furthermore, it generalizes to new unseen locations of the objects, i.e., not seen during training, in the given environments with high success rates. When compared to the state-of-the-art constrained motion planning algorithms, CoMPNet outperforms by order of magnitude improvement in computational speed with a significantly lower variance.
Reliable real-time planning for robots is essential in today's rapidly expanding automated ecosystem. In such environments, traditional methods that plan by relaxing constraints become unreliable or slow-down for kinematically constrained robots. This paper describes the algorithm Dynamic Motion Planning Networks (Dynamic MPNet), an extension to Motion Planning Networks, for non-holonomic robots that address the challenge of real-time motion planning using a neural planning approach. We propose modifications to the training and planning networks that make it possible for real-time planning while improving the data efficiency of training and trained models' generalizability. We evaluate our model in simulation for planning tasks for a non-holonomic robot. We also demonstrate experimental results for an indoor navigation task using a Dubins car.
Soft Microrobotic Transmissions Enable Rapid Ground-Based Locomotion
Wei Zhou and Nick Gravish
In this paper we present the design, fabrication, testing, and control of a 0.4 g milliscale robot employing a soft polymer flexure transmission for rapid ground movement. The robot was constructed through a combination of two methods: smart-composite-manufacturing (SCM) process to fabricate the actuators and robot chassis, and silicone elastomer molding and casting to fabricate a soft flexure transmission. We actuate the flexure transmission using two customized piezoelectric (PZT) actuators that attach to the transmission inputs. Through high- frequency oscillations the actuators are capable of exciting vibrational resonance modes of the transmission which result in motion amplification on the transmission output. Directional spines on the transmission output generate traction force with the ground and drive the robot forward. By varying the excitation frequency of the soft transmission we can control running speed, and when the transmission is oscillated at its resonance frequency we achieve high speeds with a peak speed of 439 mm/s (22 body lengths/s). By exciting traveling waves through the soft transmission, we were able to control the steer- ing direction. Overall this paper demonstrates the feasibility of exciting resonance behavior in millimeter scale soft robotic structures to achieve high-speed controllable locomotion.
Knuckles that buckle: compliant underactuated limbs with joint hysteresis enable minimalist terrestrial robots
Mingsong Jiang, Rongzichen Song and Nick Gravish
Underactuated designs of robot limbs can enable these systems to passively adapt their joint configuration in response to external forces. Passive adaptation and reconfiguration can be extremely beneficial in situations where manipulation or locomotion with complex substrates is required. A common design for underactuated systems often involves a single tendon that actuates multiple rotational joints, each with a torsional elastic spring resisting bending. However, a challenge of using those joints for legged locomotion is that limbs typically need to follow a cyclical trajectory so that feet can alternately be engaged in stance and swing phases. Such trajectories present challenges for linearly elastic underactuated limbs. In this paper, we present a new method of underactuated limb design which incorporates hysteretic joints that change their torque response during loading and unloading. A double-jointed underactuated limb with both linear and hysteretic joints can thus be tuned to create a variety of looped trajectories. We fabricate these joints inside a flexible legged robot using a modified laminate based 3D printing method, and the result shows that with passive compliance and a mechanically determined joint sequence, a 2-legged minimalist robot can successfully walk through a confined channel over uneven substrates.
OrcVIO: Object residual constrained Visual-Inertial Odometry
Mo Shan, Qiaojun Feng, Nikolay Atanasov
Introducing object-level semantic information into simultaneous localization and mapping (SLAM) system is critical. It not only improves the performance but also enables tasks specified in terms of meaningful objects. This work presents OrcVIO, for visual-inertial odometry tightly coupled with tracking and optimization over structured object models. OrcVIO differentiates through semantic feature and bounding-box reprojection errors to perform batch optimization over the pose and shape of objects. The estimated object states aid in real-time incremental optimization over the IMU-camera states. The ability of OrcVIO for accurate trajectory estimation and large-scale object-level mapping is evaluated using real data.
Dense Incremental Metric-Semantic Mapping via Sparse Gaussian Process Regression
Ehsan Zobeidi, Alec Koppel and Nikolay Atanasov
We develop an online probabilistic metric- semantic mapping approach for autonomous robots relying on streaming RGB-D observations. We cast this problem as a Bayesian inference task, requiring encoding both the geo- metric surfaces and semantic labels (e.g., chair, table, wall) of the unknown environment. We propose an online Gaussian Process (GP) training and inference approach, which avoids the complexity of GP classification by regressing a truncated signed distance function representation of the regions occupied by different semantic classes. Online regression is enabled through sparse GP approximation, compressing the training data to a finite set of inducing points, and through spatial domain partitioning into an Octree data structure with overlapping leaves. Our experiments demonstrate the effectiveness of this technique for large-scale probabilistic metric-semantic mapping of 3D environments. A distinguishing feature of our approach is that the generated maps contain full continuous distributional information about the geometric surfaces and semantic labels, making them appropriate for uncertainty-aware planning.
Fully Convolutional Geometric Features for Category-level Object Alignment
Qiaojun Feng and Nikolay Atanasov
This paper focuses on pose registration of different object instances from the same category. This is required in online object mapping because object instances detected at test time usually differ from the training instances. Our approach transforms instances of the same category to a normalized canonical coordinate frame and uses metric learning to train fully convolutional geometric features. The resulting model is able to generate pairs of matching points between the instances, allowing category-level registration. Evaluation on both synthetic and real-world data shows that our method provides robust features, leading to accurate alignment of instances with different shapes.
Estimating relative camera poses from consecutive frames is a fundamental problem in visual odometry (VO) and simultaneous localization and mapping (SLAM), where classic methods consisting of hand-crafted features and sampling-based outlier rejection have been a dominant choice for over a decade. Although multiple works propose to replace these modules with learning-based counterparts, most have not yet been as accurate, robust and generalizable as conventional methods. In this paper, we design an end-to-end trainable framework consisting of learnable modules for detection, feature extraction, matching and outlier rejection, while directly optimizing for the geometric pose objective. We show both quantitatively and qualitatively that pose estimation performance may be achieved on par with the classic pipeline. Moreover, we are able to show by end-to-end training, the key components of the pipeline could be significantly improved, which leads to better generalizability to unseen datasets compared to existing learning-based methods.
Jacobs School of Engineering