Masayoshi Tomizuka

University of California, Berkeley | UCB · Department of Mechanical Engineering

This paper proposes a path planning solution for a two-body vehicle system during complex maneuvers. The presented approach employs a kinematics-based planner to generate an initial path, which is then refined using optimization-based path smoothing techniques. The generated smooth path is reliable to be used as single source of trajectory optimiza...

I. INTRODUCTION Series elastic actuators (SEAs) have gained attention in human assist devices [1]. This study focuses on nonlinear SEAs which have the potential to overcome the performance limit of fixed stiffness SEAs[2], designs a spring unit that fits the human joint shape, and experimentally validates the torque accuracy of the kinematic model...

Obtaining accurate and diverse human motion prediction is essential to many industrial applications, especially robotics and autonomous driving. Recent research has explored several techniques to enhance diversity and maintain the accuracy of human motion prediction at the same time. However, most of them need to define a combined loss, such as the...

Motion forecasting plays a significant role in various domains (e.g., autonomous driving, human-robot interaction), which aims to predict future motion sequences given a set of historical observations. However, the observed elements may be of different levels of importance. Some information may be irrelevant or even distracting to the forecasting i...

As more robots are implemented for contact-rich tasks, tactile sensors are in increasing demand. For many circumstances, the contact is required to be compliant, and soft sensors are in need. This paper introduces a novelly designed soft sensor that can simultaneously estimate the contact force and contact location. Inspired by humans' skin, which...

Proposing grasp poses for novel objects is an essential component for any robot manipulation task. Planning six degrees of freedom (DoF) grasps with a single camera, however, is challenging due to the complex object shape, incomplete object information, and sensor noise. In this paper, we present a 6-DoF contrastive grasp proposal network (CGPN) to...

3D point-cloud-based perception is a challenging but crucial computer vision task. A point-cloud consists of a sparse, unstructured, and unordered set of points. To understand a point-cloud, previous point-based methods, such as PointNet++, extract visual features through hierarchically aggregation of local features. However, such methods have seve...

Reward function, as an incentive representation that recognizes humans' agency and rationalizes humans' actions, is particularly appealing for modeling human behavior in human-robot interaction. Inverse Reinforcement Learning is an effective way to retrieve reward functions from demonstrations. However, it has always been challenging when applying...

Generating diverse and comprehensive interacting agents to evaluate the decision-making modules of autonomous vehicles~(AV) is essential for safe and robust planning. Due to efficiency and safety concerns, most researchers choose to train adversary agents in simulators and generate test cases to interact with evaluated AVs. However, most existing m...

Autonomous vehicles (AVs) need to share the road with multiple, heterogeneous road users in a variety of driving scenarios. It is overwhelming and unnecessary to carefully interact with all observed agents, and AVs need to determine whether and when to interact with each surrounding agent. In order to facilitate the design and testing of predic...

Many manipulation tasks require robots to interact with unknown environments. In such applications, the ability to adapt the impedance according to different task phases and environment constraints is crucial for safety and performance. Although many approaches based on deep reinforcement learning (RL) and learning from demonstration (LfD) have bee...

As more and more autonomous vehicles (AVs) are being deployed on public roads, designing socially compatible behaviors for them is important. In order to generate safe and efficient actions, AVs need to not only predict the future behaviors of other traffic participants, but also be aware of the uncertainties associated with such behavior predictio...

In this contribution, we suggest two proposals to achieve fast, real-time lane-keeping control for Autonomous Ground Vehicles (AGVs). The goal of lane-keeping is to orient and keep the vehicle within a given reference path using the front wheel steering angle as the control action for a specific longitudinal velocity. While nonlinear models can des...

In this paper, a practical fractional-order variablegain super-twisting algorithm (PFVSTA) is proposed to improve the tracking performance of wafer stages for semiconductor manufacturing. Based on sliding mode control (SMC), the proposed PFVSTA enhances the tracking performance from three aspects: 1) alleviating the chattering phenomenon via super-...

An effective understanding of the environment and accurate trajectory prediction of surrounding dynamic obstacles are indispensable for intelligent mobile systems (e.g. autonomous vehicles and social robots) to achieve safe and high-quality planning when they navigate in highly interactive and crowded scenarios. Due to the existence of frequent int...

Many manipulation tasks require robots to interact with unknown environments. In such applications, the ability to adapt the impedance according to different task phases and environment constraints is crucial for safety and performance. Although many approaches based on deep reinforcement learning (RL) and learning from demonstration (LfD) have bee...

In this paper, a practical fractional-order variable-gain super-twisting algorithm (PFVSTA) is proposed to improve the tracking performance of wafer stages for semiconductor manufacturing. Based on the sliding mode control (SMC), the proposed PFVSTA enhances the tracking performance from three aspects: 1) alleviating the chattering phenomenon via s...

The precise motion control of a multi-degree of freedom~(DOF) robot manipulator is always challenging due to its nonlinear dynamics, disturbances, and uncertainties. Because most manipulators are controlled by digital signals, a novel higher-order sliding mode controller in the discrete-time form with time delay estimation is proposed in this paper...

Unlike popular modularized framework, end-to-end autonomous driving seeks to solve the perception, decision and control problems in an integrated way, which can be more adapting to new scenarios and easier to generalize at scale. However, existing end-to-end approaches are often lack of interpretability, and can only deal with simple driving tasks...

When robots work with humans for collaborative task, they need to plan their actions while taking humans' actions into account. However, due to the complexity of the tasks and stochastic nature of human collaborators, it is quite challenging for the robot to efficiently collaborate with the humans. To address this challenge, in this paper, we first...

To obtain precise motion control of wafer stages, an adaptive neural network and fractional-order super-twisting control strategy is proposed. Based on sliding mode control (SMC), the proposed controller aims to address two challenges in SMC: 1) reducing the chattering phenomenon, and 2) attenuating the influence of model uncertainties and disturba...

Peg-in-hole assembly is a challenging contact-rich manipulation task. There is no general solution to identify the relative position and orientation between the peg and the hole. In this paper, we propose a novel method to classify the contact poses based on a sequence of contact measurements. When the peg contacts the hole with pose uncertainties,...

Autonomous vehicles need to handle various traffic conditions and make safe and efficient decisions and maneuvers. However, on the one hand, a single optimization/sampling-based motion planner cannot efficiently generate safe trajectories in real time, particularly when there are many interactive vehicles near by. On the other hand, end-to-end lear...

Autonomous vehicles (AVs) need to interact with other traffic participants who can be either cooperative or aggressive, attentive or inattentive. Such different characteristics can lead to quite different interactive behaviors. Hence, to achieve safe and efficient autonomous driving, AVs need to be aware of such uncertainties when they plan their o...

The availability of many real-world driving datasets is a key reason behind the recent progress of object detection algorithms in autonomous driving. However, there exist ambiguity or even failures in object labels due to error-prone annotation process or sensor observation noise. Current public object detection datasets only provide deterministic...

Zero phase error tracking (ZPET) control has gained popularity as a simple yet effective feedforward control method for tracking time varying desired trajectories by the plant output. In this paper, we will show that the zero-order hold equivalent of continuous time transfer function, i.e. pulse transfer function, naturally has a property to realiz...

Classical game-theoretic approaches for multi-agent systems in both the forward policy design problem and the inverse reward learning problem often make strong rationality assumptions: agents perfectly maximize expected utilities under uncertainties. Such assumptions, however, substantially mismatch with observed humans’ behaviors such as satisficin...

Safety is an important topic in autonomous driving since any collision may cause serious damage to people and the environment. Hamilton-Jacobi (HJ) Reachability is a formal method that verifies safety in multi-agent interaction and provides a safety controller for collision avoidance. However, due to the worst-case assumption on the car's future ac...

We present Sparse R-CNN, a purely sparse method for object detection in images. Existing works on object detection heavily rely on dense object candidates, such as $k$ anchor boxes pre-defined on all grids of image feature map of size $H\times W$. In our method, however, a fixed sparse set of learned object proposals, total length of $N$, are provi...

A human-centered robot needs to reason about the cognitive limitation and potential irrationality of its human partner to achieve seamless interactions. This paper proposes an anytime game-theoretic planner that integrates iterative reasoning models, partially observable Markov decision process, and chance-constrained Monte-Carlo belief tree search...

Autonomous vehicles (AVs) need to share the road with multiple, heterogeneous road users in a variety of driving scenarios. It is overwhelming and unnecessary to carefully interact with all observed agents, and AVs need to determine whether and when to interact with each surrounding agent. In order to facilitate the design and testing of prediction...

LiDAR point-cloud segmentation is an important problem for many applications. For large-scale point cloud segmentation, the de facto method is to project a 3D point cloud to get a 2D LiDAR image and use convolutions to process it. Despite the similarity between regular RGB and LiDAR images, we are the first to discover that the feature distribution...

As more and more autonomous vehicles (AVs) are being deployed on public roads, designing socially compatible behaviors for them is of critical importance. Based on observations, AVs need to predict the future behaviors of other traffic participants, and be aware of the uncertainties associated with such prediction so that safe, efficient, and human...

Autonomous vehicles (AVs) need to interact with other traffic participants who can be either cooperative or aggressive, attentive or inattentive. Such different characteristics can lead to quite different interactive behaviors. Hence, to achieve safe and efficient autonomous driving, AVs need to be aware of such uncertainties when they plan their o...

Autonomous driving planning is a challenging problem when the environment is complicated. It is difficult for the planner to find a good trajectory that navigates autonomous cars safely with crowded surrounding vehicles. To solve this complicated problem, a fast algorithm that generates a high-quality, safe trajectory is necessary. Constrained Iter...

Multi-agent interacting systems are prevalent in the world, from pure physical systems to complicated social dynamic systems. In many applications, effective understanding of the situation and accurate trajectory prediction of interactive agents play a significant role in downstream tasks, such as decision making and planning. In this paper, we pro...

Safety is an important requirement in rehabilitation assist suits. We have developed a velocity-based mechanical safety device (VBMSD) for an assist suit to aid in the flexion and extension of a patient’s knee joint. The VBMSD is attached to the assist suit. The VBMSD stops the suit’s motor when the angular velocity of the knee joint matches or exc...

Iterative learning control (ILC) can synthesize the feedforward control signal for the trajectory tracking control of a repetitive task, even when the system has strong nonlinear dynamics. This makes ILC be one of the most popular methods for trajectory tracking control. Restriction on a repetitive task, however, limits its application to multiple...

In human-robot interaction (HRI) systems, such as autonomous vehicles, understanding and representing human behavior are important. Human behavior is naturally rich and diverse. Cost/reward learning, as an efficient way to learn and represent human behavior, has been successfully applied in many domains. Most of traditional inverse reinforcement le...

The author of this article Jun Moon has changed the affiliation.

In this paper, a novel fractional-order variable-gain super-twisting control (FVSTC) scheme is proposed and applied to improve the tracking performance of wafer stages in the photolithography systems. The FVSTC overcomes the drawbacks of the super-twisting control (STC) such as slow response speed and incomplete compensation to disturbances. First,...

Abstract—Individuals with neurological impairment, particularly those with cervical level spinal cord injuries (SCI), often have difficulty with daily tasks due to triceps weakness or total loss of function. More demanding tasks, such as sit-skiing, may be rendered impossible due to their extreme strength demands. Design of exoskeletons that addres...

In the past decades, we have witnessed significant progress in the domain of autonomous driving. Advanced techniques based on optimization and reinforcement learning become increasingly powerful when solving the forward problem: given designed reward/cost functions, how we should optimize them and obtain driving policies that interact with the envi...

In this paper, we propose a novel form of the loss function to increase the performance of LiDAR-based 3D object detection and obtain more explainable and convincing uncertainty for the prediction. The loss function was designed using corner transformation and uncertainty modeling. With the new loss function, the performance of our method on the va...

In this paper, we propose a novel form of the loss function to increase the performance of LiDAR-based 3d object detection and obtain more explainable and convincing uncertainty for the prediction. The loss function was designed using corner transformation and uncertainty modeling. With the new loss function, the performance of our method on the va...

In the past decades, we have witnessed significant progress in the domain of autonomous driving. Advanced techniques based on optimization and reinforcement learning (RL) become increasingly powerful at solving the forward problem: given designed reward/cost functions, how should we optimize them and obtain driving policies that interact with the e...

Rectifier (ReLU) deep neural networks (DNN) and their connection with piecewise affine (PWA) functions is analyzed. The paper is an effort to find and study the possibility of representing explicit state feedback policy of model predictive control (MPC) as a ReLU DNN, and vice versa. The complexity and architecture of DNN has been examined through...

Computer vision has achieved great success using standardized image representations -- pixel arrays, and the corresponding deep learning operators -- convolutions. In this work, we challenge this paradigm: we instead (a) represent images as a set of visual tokens and (b) apply visual transformers to find relationships between visual semantic concep...

Reinforcement learning methods have been developed to achieve great success in training control policies in various automation tasks. However, a main challenge of the wider application of reinforcement learning in practical automation is that the training process is hard and the pretrained policy networks are hardly reusable in other similar cases....

Base on the accurate inverse of a system, the feedforward compensation method can compensate the tracking error of a linear system dramatically. However, many control systems have complex dynamics and their accurate inverses are difficult to obtain. In the paper, a variable parameter model is proposed to describe a system and a multi-step adaptive...

In this paper, a method for nonlinear control is proposed using an extended state observer for position tracking of an electro-hydraulic system (EHS) with only position feedback. The proposed method consists of a high gain extended state observer (HGESO) and a nonlinear controller. The EHS model is transformed into the normal form to lump a system...

Trajectory planning is of vital importance to decision-making for autonomous vehicles. Currently, there are three popular classes of cost-based trajectory planning methods: sampling-based, graph-search-based, and optimization-based. However, each of them has its own shortcomings, for example, high computational expense for sampling-based methods, l...

Due to an unfortunate oversight the acknowledgment has been omitted:

Accurately predicting the possible behaviors of traffic participants is an essential capability for autonomous vehicles. Since autonomous vehicles need to navigate in dynamically changing environments, they are expected to make accurate predictions regardless of where they are and what driving circumstances they encountered. A number of methodologi...

LiDAR point-cloud segmentation is an important problem for many applications. For large-scale point cloud segmentation, the \textit{de facto} method is to project a 3D point cloud to get a 2D LiDAR image and use convolutions to process it. Despite the similarity between regular RGB and LiDAR images, we discover that the feature distribution of LiDA...

Current autonomous driving systems are composed of a perception system and a decision system. Both of them are divided into multiple subsystems built up with lots of human heuristics. An end-to-end approach might clean up the system and avoid huge efforts of human engineering, as well as obtain better performance with increasing data and computatio...

The availability of real-world datasets is the pre-requisite to develop object detection methods for autonomous driving. While ambiguity exists in object labels due to error-prone annotation process or sensor observation noises, current object detection datasets only provide deterministic annotations, without considering their uncertainty. This pre...

Human-robot collaboration (HRC) is becoming increasingly important as the paradigm of manufacturing is shifting from mass production to mass customization. The introduction of HRC can significantly improve the flexibility and intelligence of automation. To efficiently finish tasks in HRC systems, the robots need to not only predict the future movem...

The aim of this study is to examine the proposed control method of the assist suit with a Velocity-Based Mechanical Safety Device (VBMSD) for patients with difficulty moving their lower legs by themselves. The proposed control method for the assist suit assists the patients as if the patients move their knee joint under zero gravity. A physical sim...

Human motion prediction is non-trivial in modern industrial settings. Accurate prediction of human motion can not only improve efficiency in human robot collaboration, but also enhance human safety in close proximity to robots. Among existing prediction models, the parameterization and identification methods of those models vary. It remains unclear...

Unlike popular modularized framework, end-to-end autonomous driving seeks to solve the perception, decision and control problems in an integrated way, which can be more adapting to new scenarios and easier to generalize at scale. However, existing end-to-end approaches are often lack of interpretability, and can only deal with simple driving tasks...

Reinforcement learning methods have been developed to achieve great success in training control policies in various automation tasks. However, a main challenge of the wider application of reinforcement learning in practical automation is that the training process is hard and the pretrained policy networks are hardly reusable in other similar cases....

To obtain precise motion control of wafer stages, an adaptive neural network and fractional-order super-twisting control strategy is proposed. Based on sliding mode control (SMC), the proposed controller aims to address two challenges in SMC: 1) reducing the chattering phenomenon, and 2) attenuating the influence of model uncertainties and disturba...

Sampling-based motion planning methods are widely adopted in autonomous driving. Typically, sampling can be decoupled into two layers: a path sampling layer and a speed profile sampling layer. For the path sampling layer, traditional methods tend to sample with a uniform distribution over the whole feasible space, which might cause either computati...

This paper presents a robust probabilistic point registration method for estimating the rigid transformation (i.e. rotation matrix and translation vector) between two pointcloud dataset. The method improves the robustness of point registration and consequently the robot localization in the presence of outliers in the pointclouds which always occurs...

Mapping and localization is a critical module of autonomous driving, and significant achievements have been reached in this field. Beyond Global Navigation Satellite System (GNSS), research in point cloud registration, visual feature matching, and inertia navigation has greatly enhanced the accuracy and robustness of mapping and localization in dif...

Crowd counting in images is a widely explored but challenging task. Though recent convolutional neural network (CNN) methods have achieved great progress, it is still difficult to accurately count and even to precisely localize people in very dense regions. A major issue is that dense regions usually consist of many instances of small size, and thu...

The neural network policies are widely explored in the autonomous driving field, thanks to their capability of handling complicated driving tasks. However, the practical deployment of such policies is slowed down due to their lack of robustness against modeling gap and external disturbances. In our prior work, we proposed a planner-controller archi...

Probabilistic vehicle trajectory prediction is essential for robust safety of autonomous driving. Current methods for long-term trajectory prediction cannot guarantee the physical feasibility of predicted distribution. Moreover, their models cannot adapt to the driving policy of the predicted target human driver. In this work, we propose to overcom...

Applying the maximum correntropy criterion (MCC) to multi-frame alignment problem, the objective function is as following where denotes the squared point-to-plane distance, denotes the correspondence pair indicating whether two LiDAR measurements have enough spatial overlapping. ➢ Iterative Reweighted Least Squares (IRLS): Since correntropy metric...