Evolutionary Decision-Making and Planning for Autonomous Driving Based on Safe and Rational Exploration and Exploitation

Decision-making and motion planning are extremely important in autonomous driving to ensure safe driving in a real-world environment.This study proposes an online evolutionary decision-making and motion planning framework for autonomous driving based on a hybrid data-and model-driven method.First,a data-driven decision-making module based on deep reinforcement learning(DRL)is developed to pursue a rational driving performance as much as possible.Then,model predictive control(MPC)is employed to execute both longitudinal and lateral motion planning tasks.Multiple constraints are defined according to the vehicle’s physical limit to meet the driving task requirements.Finally,two principles of safety and rationality for the self-evolution of autonomous driving are proposed.A motion envelope is established and embedded into a rational exploration and exploitation scheme,which filters out unreasonable experiences by masking unsafe actions so as to collect high-quality training data for the DRL agent.Experiments with a high-fidelity vehicle model and MATLAB/Simulink co-simulation environment are conducted,and the results show that the proposed online-evolution framework is able to generate safer,more rational,and more efficient driving action in a real-world environment.

A Novel Approach for Evaluation of Food Functions and Safety Applied in RR GM Soybeans

[Objective] This study aimed to evaluate tbe healthy risk of genetically modified （ GM ） soybeans by using a novel approach for functions and safety of food. [ Me^od] Different from traditional evaluation of substantial equivalence, three great innovations were performed in this study, involving in basic diet, evalu- ation approaches and principle, as well as the clarification of connotation differences between absolute and relative mass of organs. Hence a novel BDI-GS （Bendib Damage Index and General Score） evaluation approach was established and applied in comparative evaluation between RR GM and natural soybeans. Healthy male ICR mice during linear growth were selected; experimental mice were fed with 15% RR GM soybeans and 15% natural soybeans blending maize meal diets, and control mice were fed with single maize meal diet for 13 d; the mice were dissected after collecting blood samples and perfectly obtained nine organs or tissues to re- cord their masses and conduct statistical analyses. [Result] Plenty of matching information was obtained through simple design. The growth performance of treated mice was markedly of individual differences, some mice were thwarted due to regular intake of RR soybeans. Meanwhile, the functions and safety of RR soybeans were markedly lowered in overall nutritional and healthy effects than those of natural soybeans expressed in GS values, and presents some declines in nutrition and health of thymus, pancreas and spermary; especially, it can make thymus immune （P 〈0.05） in markedly lower level than that of natural soybeans. [ Conclusion] Therefore, major troubles and risks of RR soybeans intake are of personal risks in different degrees, in addition, it may increase sub-health and related chronic epi- demics risks, and herein it will presents certain safety issues. The creation of this novel evaluation system provides a simple and available evaluation approach for functions and potential risks revelation of food effects, and will yield far-reaching influences to safety evaluation and healthy development of GM foods, as well as public health.