Mitsubishi Electric Research Laboratories, Inc.

Mitsubishi Electric Corporation

Karl Berntorp

Stefano Di Cairano

Marcel Menner

Marcus Greiff

Philip Orlik

Avishai Weiss

Pu Wang

Rien Quirynen

Petros T Boufounos

Hassan Mansour

Ankush Chakrabarty

Claus Danielson

Kieran Parsons

Yebin Wang

Kyeong Jin Kim

Kyeong-Jin Kim

Gang Yao

Toshiaki Koike-Akino

Ming-Yu Liu

Jianlin Guo

Uros Kalabic

Heejin Ahn

Abraham P Vinod

Eyad Zeino

Toshiaki Akino

Shuangxuan Ran

Astha Tiwari

Terrence Skibik

Sixiong You

Remy Zawislak

Hassan Iqbal

Stefano Di Calrano

Ferdous Pervej

Ziyi Ma

Peizhao Li

Yuxuan Xia

Sean Vaskov

Uros Kalabić

Oktay Arslan

Kazuhide Okamoto

Stefano Dl Cairano

A control system for controlling a motion of an ego-vehicle traveling to a target destination is provided. The control system includes a memory and a processor to execute instruction stored by the memory. The memory stores multiple trajectory-generating functions corresponding to a maneuver defined by a parameter vector associated with a driving decision. The parameter vector is defined by one or multiple parameters. Each of the multiple trajectory-generating functions is configured to generate an achieving sequence of regions of states and values of the parameter vector reaching an input target region within a prediction horizon. The stored instructions cause the control system to test control admissibility of at least some of the driving decisions consistent with the target destination of the ego-vehicle at a current state. The stored instruction also caused the control system to control the ego-vehicle according to one of the admissible driving decisions.

Systems and methods for decision making for autonomous vehicles

A tracking system for tracking an expanded state of an object is provided. The tracking system executes, for a predetermined time period, a probabilistic filter that iteratively tracks a belief on the expanded state of the object, wherein the belief is predicted using a motion model of the object and is further updated using a compound measurement model of the object. After the predetermined time period, the updated beliefs are smoothed to generate a state-decoupled online batch of training data. The compound measurement model includes multiple probabilistic distributions constrained to lie around a contour of the object with a predetermined relative geometrical mapping to the center of the object. The compound measurement model is updated using the online batch of training data. Further, the tracking system tracks the expanded state of the object based on the updated compound measurement model.

System and method for tracking an expanded state of a moving object using an online adapted compound measurement model

A computer-implemented method is provided for training a global machine learning model using a learning server and a set of vehicle agents connected to roadside units (RSUs). It involves selecting agents, associating them with RSUs based on proximity, and transmitting training data and deadlines. Agents train models locally, which are then aggregated through RSUs to refine the global model to desired precision. The method includes steps of selecting vehicle agents from a pool of the vehicle agents connected to the RSUs, associating the selected vehicle agents and the RSUs respectively based on distances from the selected vehicle agents to the RSUs configured to provide measurements of the distances to the learning server, and transmitting a global model, a selected agent set and deadline thresholds in each global training round to the RSUs configured to transmit the global model and training deadlines to the selected vehicle agents.

Communication and computation aware distributed machine learning for vehicular networks

The present disclosure discloses a system and a method for controlling a vehicle in an environment. The method uses a processor coupled to a memory storing a probabilistic map of the environment relating measurements of a magnetic field and time of the measurements of the magnetic field to a probability of locations within the environment. The processor is coupled with stored instructions when executed by the processor carry out steps of the method comprising estimating a probability of the current location in the environment based on a current measurement of a magnetometer at a current timestamp by submitting the current measurement and the current timestamp to the probabilistic map, and controlling an actuator of the vehicle based on a stochastic control employing the probability of the current location.

System and method for controlling a vehicle in an environment

The present disclosure provides a feedback controller and method for controlling an operation of a device at different control steps. The feedback controller comprises at least one processor, and the memory having instructions stored thereon that, when executed by the at least one processor, causes the feedback controller, for a control step, to collect a measurement indicative of a state of the device at the control step, and execute, recursively until a termination condition is met, a probabilistic solver parameterized on a control input to an actuator operating the device to produce a control input for the control step. The feedback controller is further configured to control the actuator operating the device based on the produced control input.

System and method for controlling an operation of a device

The present disclosure discloses a system and a method for controlling a device using a compound probabilistic filter. The method comprises collecting a sequence of measurements indicative of the state of the device at different control steps. Further, the method comprises executing iteratively a compound probabilistic filter configured to track the state of the device at each of the different control steps using the sequence of measurements to produce a sequence of states of the device corresponding to the sequence of measurements. Furthermore, the method comprises controlling the device using the tracked state of the device.

System and method for controlling a device using a compound probabilistic filter

Embodiments of the present disclosure disclose a method and a system for tracking the position of the one or more moving objects. The method includes collecting GNSS measurement data of satellite signals transmitted from multiple satellites. The method further includes extracting values of a plurality of features from the GNSS measurement data. The method includes mapping the extracted values of the plurality of features to a source domain. The method includes classifying the mapped transformed plurality of features using a neural network. The neural network is trained over simulated data sampled from a source domain. The method includes identifying multipath measurements of the GNSS measurement data based on classification of the corresponding mapped transformed plurality of features. The method includes tracking the position of the one or more moving objects by processing identification of GNSS measurement data affected by multipath.

Method and system for position estimation using domain adaptation

A system for controlling motion of an articulated vehicle with one or multiple trailers is described. The system is configured to infer the state of a vehicle, determine a motion path with forward motions and backward motions for the vehicle based on the state, and formulate an optimal control problem for optimizing the motion path. The optimal control problem comprises an integral tracking error function indicating integral tracking error based on motion cusps relating to the motion path. The motion cusps indicate switching between forward motion and backward motion. The motion path is optimized over a prediction horizon based on solving the optimal control problem. A control command is generated for the vehicle based on the optimized motion path and a vehicle model, and the motion of the articulated vehicle is controlled based on the control command thereby changing its state.

System and method for controlling motion of articulated vehicles

A controller is provided for operating a system under admissible states. The controller includes an interface configured to connect the system storing a set of measured system states, a set of reference inputs and a set of system parameters in a storage arranged inside or outside the system, a memory storing measured system states, admissible reference inputs and admissible parameter sets and computer-executable programs including a parameter estimator and an adaptive reference governor (ARG), a processor, in connection with the memory. The processor is configured to perform the ARG and the parameter estimator. The parameter estimator extracts a pair of a reference input and the system state and compute a system parameter estimate based on the reference input and system state. The ARG is configured to update the reference input and compute a parameter-robust constraint admissible set based on the updated reference input and the system states, wherein the ARG generates and transmits a reference input to the system based on the parameter-robust constraint admissible set.

System and method for data-driven reference generation

The present disclosure provides a system and a method for detecting and tracking objects. The method includes permuting an order of frames in a sequence of radar image frames to produce multiple permuted sequences with different frames at a dominant position in a corresponding permuted sequence of radar image frames. Each permuted sequence of radar image frames is processed with a first neural network to produce temporally enhanced features for each of the frames in the sequence of radar image frames. Further a feature map is reconstructed from the temporally enhanced features of each of the frames in the sequence of radar image frames to produce a sequence of feature maps. The method further includes processing a list of feature vectors from each feature map with a second neural network to produce temporally enhanced heatmaps.

System and method for radar object recognition with cross-frame temporal relationality

A probabilistic feedback controller for controlling an operation of a robotic system using a probabilistic filter subject to a structural constraint on an operation of the robotic system is configured to execute a probabilistic filter estimates a distribution of a current state of the robotic system given a previous state of the robotic system based on a motion model of the robotic system perturbed by stochastic process noise and a measurement model of the robotic system perturbed by stochastic measurement noise having an uncertainty modeled as a time-varying Gaussian process represented as a weighted combination of time-varying basis functions with weights defined by corresponding Gaussian distributions. The probabilistic filter recursively updates both the distribution of the current state of the robotic system and the Gaussian distributions of the weights of the basis functions selected to satisfy the structural constraint indicated by measurements of the state of a robotic system.

Model-based control with uncertain motion model

A vehicle includes an advanced driver-assistance system (ADAS) configured to intervene in an operation of the control system by complementing or overriding the driving input in response to detecting a driving condition dependent on a calibration parameter indicative of a preference of execution of the driving maneuver. The ADAS is calibrated based on a crowd-local distribution function of the calibration parameter indicative of a distribution of the preference of execution of the driving maneuver by other drivers of other vehicles at a specific location or a specific environment in response to detecting that the vehicle approaches the specific location or the specific environment.

Method and system for driving condition-agnostic adaptation of advanced driving assistance systems

A system for controlling an operation of a machine for performing a task is disclosed. The system submits a sequence of control inputs to the machine and receives a feedback signal. The system further determines, at each control step, a current control input for controlling the machine based on the feedback signal including a current measurement of a current state of the system by applying a control policy transforming the current measurement into the current control input based on current values of control parameters in a set of control parameters of a feedback controller. Furthermore, the system may iteratively update a state of the feedback controller defined by the control parameters using a prediction model predicting values of the control parameters and a measurement model updating the predicted values to produce the current values of the control parameters that explain the sequence of measurements according to a performance objective.

System and method for calibrating feedback controllers

A server jointly tracks states of multiple vehicles using measurements of satellite signals received at each vehicle and parameters of the probabilistic distribution of the state of each vehicle. The server fuse states and measurements into an augmented state of the multiple vehicles and an augmented measurement of the augmented state subject to augmented measurement noise defined by a non-diagonal covariance matrix with non-zero off-diagonal elements, each non-zero off-diagonal elements relating errors in the measurements of a pair of corresponding vehicles. The server executes a probabilistic filter updating the augmented state and fuses the state of at least some of the multiple vehicles with a corresponding portion of the updated augmented state.

Cooperative state tracking of multiple vehicles using individual and joint estimations

Stochastic nonlinear model predictive control (SNMPC) allows to directly take uncertainty of the dynamics and/or of the system's environment into account, e.g., by including probabilistic chance constraints. However, SNMPC requires the approximate computation of the probability distributions for the state variables that are propagated through the nonlinear system dynamics. This invention proposes the use of Gaussian-assumed density filters (ADF) to perform high-accuracy propagation of mean and covariance information of the state variables through the nonlinear system dynamics, resulting in a tractable SNMPC approach with improved control performance. In addition, the use of a matrix factorization for the covariance matrix variables in the constrained optimal control problem (OCP) formulation guarantees positive definiteness of the full trajectory of covariance matrices in each iteration of any optimization algorithm. Finally, a tailored adjoint-based sequential quadratic programming (SQP) algorithm is described that considerably reduces the computational cost and allows a real-time feasible implementation of the proposed ADF-based SNMPC method to control nonlinear dynamical systems under uncertainty.

Growing community of inventors

Watertown, MA, United States of America

Karl Berntorp