# US 2019/0378041 A1 · Predictive control techniques for ground vehicles

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0378041 A1
- Document ID: US20190378041A1
- Status: Published application
- Kind: A1
- Application number: 16/438,319
- Filed: 2019-06-11
- Published: 2019-12-12
- Earliest claimed: 2018-06-11
- Assignees: Traxen Inc.
- Applicants: Traxen Inc.
- Inventors: Naren Reddy DHANSRI; Nabil HK HIRZALLAH; Michael MAKOWSKI; Mahmood KARIMI; Gerti TUZI; Syed ALI; Saeed ZAMANZAD GAVIDEL; Ali MALEKI
- Web page: https://patentgrep.com/patent/US20190378041A1

## CPC

- B60W 30/143: {Speed control (B60W30/16 takes precedence)}
- G06N 20/00: Machine learning
- B60W 30/14: {Adaptive} cruise control
- B60W 30/18145: {Cornering}
- B60W 50/0097: {Predicting future conditions}
- G01S 13/931: of land vehicles
- G01S 17/931: of land vehicles
- G06N 3/006: based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- B60W 2050/0013: {Optimal controllers}
- B60W 2400/00: Indexing codes relating to detected, measured or calculated conditions or factors
- B60W 2420/403: Image sensing, e.g. optical camera
- B60W 2520/10: Longitudinal speed
- B60W 2520/105: Longitudinal acceleration
- B60W 2520/14: Yaw
- B60W 2520/28: Wheel speed
- B60W 2540/10: Accelerator pedal position
- B60W 2540/12: Brake pedal position
- B60W 2552/15: Road slope, i.e. the inclination of a road segment in the longitudinal direction
- B60W 2552/20: Road profile, i.e. the change in elevation or curvature of a plurality of continuous road segments
- B60W 2552/30: Road curve radius
- B60W 2552/53: Road markings, e.g. lane marker or crosswalk
- B60W 2554/80: Spatial relation or speed relative to objects
- B60W 2555/20: Ambient conditions, e.g. wind or rain
- B60W 2555/60: Traffic rules, e.g. speed limits or right of way
- B60W 2720/10: Longitudinal speed
- G01S 13/582: {adapted for simultaneous range and velocity measurements}
- G01S 13/584: {adapted for simultaneous range and velocity measurements}
- G01S 13/86: Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
- G01S 13/865: {Combination of radar systems with lidar systems}
- G01S 13/867: {Combination of radar systems with cameras}
- G01S 17/86: Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
- G01S 2013/932: {using own vehicle data, e.g. ground speed, steering wheel direction}
- G01S 2013/9321: {Velocity regulation, e.g. cruise control}
- G01S 2013/9322: {using additional data, e.g. driver condition, road state or weather data}
- G01S 2013/9323: {Alternative operation using light waves}
- G01S 2013/93271: {in the front of the vehicles}
- Y02T 10/40: Engine management systems
- B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- Y02T: CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION

## Abstract

A ground vehicle control system including a plurality of sensors and one or more predictive controllers. The sensors can be configured to detect environment information and vehicle operating information. The one or more predictive controllers can be configured to self-train for an energy consumption solution based on one or more control parameters including the environment information and the vehicle operating information.

## Claims

### Claim 1

A ground vehicle control system comprising:
a plurality of sensors configured to detect environment information and vehicle operating information;
one or more predictive controllers configured to
self-train for an energy consumption solution based on one or more control parameters including the environment information and the vehicle operating information, wherein the environment information includes one or more of current roadway topography, future roadway topography, traffic information, and traffic control device state and the vehicle operating information includes one or more of fuel map and vehicle mass; and
generate one or more control signals based on the energy consumption solution for output in a vehicle control mode selected from the group consisting of an active control mode, an assist control mode, a coaching control mode, and a passive control mode.

### Claim 2 (depends on 1)

The ground vehicle control system of claim 1, wherein the one or more predictive controllers are configured to optimize energy consumption base on:
segmenting one or more maps including creating tags of driving speeds for road segments controlling a speed of the ground vehicle within a speed window using one or more of road preview information and environment information;
scheduling shifting using one or more of the road preview information and the environment information;
planning an acceleration/deceleration profile in one or more of a traffic situation and a non-traffic situation using one or more of the road preview information and the environment information;
applying torque based on a speed-torque efficiency profile of the ground vehicle; and
determining a braking plan using one or more of road preview information and environment information.

### Claim 3 (depends on 1)

The ground vehicle control system of claim 1, wherein the one or more controllers comprise one or more model predictive controllers.

### Claim 4 (depends on 3)

The ground vehicle control system of claim 3, wherein the one or more model predictive controllers are configured to control one or more of vehicle torque, vehicle gear, vehicle retardation, vehicle speed, vehicle acceleration, and vehicle fuel consumption to minimize fuel consumption.

### Claim 5 (depends on 3)

The ground vehicle control system of claim 3, wherein the one or more model predictive controllers are configured to optimize energy consumption based on a cost function for one or more decision variables that include engine torque, axle torque, retardation torque, vehicle speed, and distance traveled, subject to a vehicle dynamics for a prediction horizon, and one or more of soft constraints and hard constraints, time varying and non-time varying.

### Claim 6 (depends on 5)

The ground vehicle control system of claim 5, wherein the optimization problem is solved by iteratively finding a search direction and a step size, where the search direction is found by simultaneously solving a set of linear equations that include a positive definite block diagonal matrix and a tridiagonal matrix.

### Claim 7 (depends on 1)

The ground vehicle control system of claim 1, wherein the one or more model predictive controllers includes:
a vehicle trajectory planning controller coupled to one or more external sensor, one or more external data sources and one or more ground vehicle sensors;
a vehicle dynamic controller coupled to the vehicle trajectory planning controller;
a vehicle stability and wheel dynamic controller coupled to one or more ground vehicle sensors; and
an actuator controller coupled to the vehicle dynamic controller, the vehicle stability and wheel dynamic controller and one or more ground vehicle sensors.

### Claim 8 (depends on 7)

The ground vehicle control system of claim 7, wherein the one or more model predictive controllers prioritize inputs to a plurality of actuators of the ground vehicle based on one or more of priorities, constraints and preview information.

### Claim 9 (depends on 1, 3)

The ground vehicle control system of claim 1, wherein the one or more controllers configured to optimize energy consumption includes:
a linear generalized predictive controller; and
successive linearization to linearize the system at sample instants to result in a quadratic programming problem, a result of the quadratic programming problem is used as control inputs for a control horizon. The ground vehicle control system of claim 3, wherein:
the one or more control parameters including one or more of weights and constraints are tuned utilizing one of machine learning, multi-object optimization or logic based techniques.

### Claim 10 (depends on 2)

The ground vehicle control system of claim 2, wherein:
the one or more controllers are further configured to optimize energy consumption based on a lead vehicle model.

### Claim 11 (depends on 14)

The ground vehicle control system of claim 14, wherein the lead vehicle model is based on one or more of a predicted future acceleration of one or more lead vehicles, a velocity of the one or more lead vehicles, a position of the one or more lead vehicles and a receding horizon control strategy.

### Claim 12 (depends on 2)

The ground vehicle control system of claim 2, further comprising one or more backup controllers, wherein control of the ground vehicle switches to control by the one or more backup controllers when energy consumption optimization is not feasible, convergence issue, in the event accurate mass or other parameters are not available or accurate, or in high deceleration required events.

### Claim 13 (depends on 12)

The ground vehicle control system of claim 12, wherein control of the ground vehicle switches back from one or more backup controllers when the one or more controllers can converge to energy consumption optimization within a prescribed iteration limit, the one or more controllers calculate a satisfactory control signal and the backup controller has been operating for a predetermined amount of time.

### Claim 14 (depends on 12)

The ground vehicle control system of claim 12, wherein the one or more backup controllers are configured to:
compensate for disturbances and plant uncertainties in the linear system; and
linear quadratic reference tracking speed and relative distance while minimizing instantaneous fuel consumption.

### Claim 15 (depends on 2)

A ground vehicle control system of claim 2, wherein the one or more predictive controllers are further configured to;
determine one or more driving scenarios of the one or more surrounding vehicles based on the one or more driving environment parameters;
tune one or more driver control inputs and one or more ground vehicle operating parameters based on the one or more driving scenarios; and
generate one or more control signals based on the one or more tuned driver control inputs and one or more tuned ground vehicle operating parameters for output in a vehicle control mode selected from the group consisting of an active control mode, an assist control mode, a coaching control mode, and a passive control mode.

### Claim 16 (depends on 15)

The ground vehicle control system of claim 15, wherein the one or more predictive controllers are further configured to:
determine a target speed window based on one or more of one or more ground vehicle operating parameters and one or more driving environment parameters; and
determine a target vehicle performance plan based on the target speed window and one or more of the one or more ground vehicle operating parameters, the driver control inputs and the one or more driving environment parameters including the at least one predicted driving environment parameter to reduce one or more of energy consumption and vehicle retardation by a ground vehicle.

### Claim 17 (depends on 16)

The ground vehicle control system of claim 16, wherein the one or more predictive controllers are further configured to:
compare the target vehicle performance plan to the one or more driving scenarios of the one or more surrounding vehicles;
classify one or more surrounding vehicles based on differences between the target vehicle performance plan and the one or more driving scenarios of the one or more surrounding vehicles; and
tune the one or more driver control inputs and the one or more ground vehicle operating parameters based on the one or more driving scenarios.

### Claim 18 (depends on 17)

The ground vehicle control system of claim 17, wherein machine learning is used to classify the surrounding vehicles as good or bad based on the one or more driving scenarios of the one or more surrounding vehicles.

### Claim 19 (depends on 17)

The ground vehicle control system of claim 17, wherein the one or more controllers configured to compare the target vehicle performance plan to the one or more driving scenarios of the one or more surrounding vehicles includes the one or more controller further configured to:
initially compare the acceleration/deceleration profile plan to a velocity and an acceleration of the one or more surrounding vehicles to initially determine a good or bad categorization for each of the one or more surrounding vehicles; and
further compare operating parameters, driver control inputs, and one or more driving environment parameters for the host vehicle to the velocity, acceleration, distance, type of vehicle signal light activation, current lane, and past behavior of the one or more surrounding vehicles to determine the good or bad categorization of each of the one or more surrounding vehicles.

### Claim 20

A method comprising:
running in a Monte Carlo loop, sampling a set of parameters from a distribution parameterized for a local environment;
running in a simulation loop within the Monte Carlo loop, performing simulations by an offline controller parameterized for the set of sampled parameters using a vehicle model and a sequence of states of the local environment;
running in the simulation loop, determining samples of losses for the simulations using the vehicle model and the sequence of states of the local environment until a stopping criterion is satisfied;
running in the Monte Carlo loop, selecting a set of given control parameters that yield a predetermined loss after the stopping criterion is satisfied; and
setting control parameters of an online controller based on the set of given control parameters that yield the predetermined loss.

### Claim 21 (depends on 20)

The method according to claim 20, wherein the predetermined loss comprises a minimum loss.

### Claim 22 (depends on 20)

The method according to claim 20, wherein the states of the local environments are based on a plurality of ground vehicles operating in the local environment.

### Claim 23 (depends on 20)

The method according to claim 20, wherein
the offline controller does not control a ground vehicle; and
the online controller does control the ground vehicle.

### Claim 24 (depends on 20)

The method according to claim 20, wherein the offline controller comprises a model predictive controller parameterized by the set of sampled model parameters.

### Claim 25 (depends on 20)

The method according to claim 20, wherein the sequence of state of the local environment comprise a temporal sequence of states generated by vehicles in the local environment.

### Claim 26 (depends on 20)

The method according to claim 20, wherein the samples of losses are determined from a loss function for the simulation.
